(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Mountain Geography A Critique and Field Study"

OSMANIA UNIVERSITY LIBRARY 

pall No. C$1 I P 3^ A/ Acc^inn N S . %jQbO 



Vuthor 
"itk 



rhis book should be returned <m or bdbu dig- (Lite I i/t iu ifkrd beK>\v. 



LONDON : HUMPHREY MILFORD 
OXFORD UNIVERSITY PRESS 



MOUNTAIN GEOGRAPHY 

A Critique and Field Study 



BY 

RODERICK PEATTIE 

PROFESSOR OF GEOGRAPHY, OHIO STATE UNIVERSITY 




Cambridge, Massachusetts 

HARVARD UNIVERSITY PRESS 

1936 



COPYRIGHT, I936 
BY THE PRESIDENT AND FELLOWS OF HARVARD COLLEGE 



PRINTED AT THE HARVARD UNIVERSITY PRESS 
CAMBRIDGE, MASS., U.S.A. 



To 



MARGARET RHODES PEATTIE 

WHO SHARED WITH ME THE SACRIFICES NECESSARY TO CARRY 

ON THE STUDY AS WELL AS THE PLEASURES 

OF VOYAGING AND RESEARCH 



PREFACE 

A MAN may live a lifetime in a group of mountains and not 
complete the study of that single group. So much is there 
to know in geology, morphology, climate, plant ecology, and 
human responses and economies, that for one volume to treat of 
the geography of the mountains of the world may seem presump- 
tuous. Also the writer's experience is, alas, more limited than he 
could wish. In America he has had experience or definite field 
work in the Laurentide and Notre Dame Mountains of Canada, 
and the Green and Taconic Mountains, the Blue Ridge, Ozarks, 
Rockies, and Sierra Nevadas of the United States. Most of his 
serious field study has been in Europe. There he has visited or 
studied the Sierra Nevada, the Sierra de Guadarrama, the Pyr- 
enees, the French Alps in Provence, Dauphine, and Savoy, and 
the Jura; in Switzerland the Valais, the Bernese Oberland, the 
St. Gotthard massif, Grisons, and Engadine; in Austria the Tirol, 
particularly the Stubaier Alps, the Salzburgkammer, the Tauren 
Alps, and the Karawankens of Carinthia; the Dalmatian Alps of 
Yugoslavia; the Trentino (Sud Tirol), the Apennines, and the 
Sicilian upland in Italy; the Bavarian Alps and the Black Forest. 
Many lesser ranges, as the Scottish Highlands, the Julian Alps, 
and the Albanian Mountains, he has seen sufficiently to judge 
their character. 

This book is a personal field study and also a review of the 
present knowledge of mountains. The European studies in the 
field are many. Not only are mountains interesting to geogra- 
phers because of the varied factors which present themselves 
within small areas, but love of mountains has called many scien- 
tists to highlands for vacation sojourns. With a few exceptions 
the European studies have a regional point of view. There is to 
date little in the way of a general critical summary. 1 The present 
is an attempt at this. The volume is, however, meant less for the 
use of Europeans than for that of Americans who have had little 
contact with the well populated mountains and who lack con- 
venient access to much of the mountain literature. 

The writer has spent time in the libraries of the Sorbonne, 

1 The recent study by Jules Blache, Vhomme et la montagne, is a notable exception. 



viii PREFACE 

Heidelberg, and Munich. The University of Grenoble, where 
Raoul Blanchard maintains the Institut de Geographie Alpine, 
offered great assistance. A second institute of this type has re- 
cently been founded at the University of Innsbruck. The writer 
has had the advantage of reading and conferences at both these 
institutions. He has also visited the special library on mountain 
geography, the Bibliotheque Raymond at Bagneres-de-Bigorre. 
The library of the Geographical Society at Madrid gave access 
to the material on Spain. He had the advantage of conferences 
with courteous scientists at Montpellier, Zurich, and Florence. 
The assistance which was freely given to a stranger at each of 
these places has made possible this study and has demonstrated 
the universal brotherhood of men interested in science. 

The writer would hesitate to weigh the mountain studies of 
others if he had not pursued at one time or another many of the 
type problems which have been investigated. He began his physio- 
graphic studies in mountains as a student with W. W. Atwood 
in the San Juan Range in 191 1. His first climatic study of 
mountains was made under the late Robert DeCourcy Ward 
at Harvard University in 1916. In 1927-28 and 1931-32 
he made a number of personal field studies in Sicily, Spain, 
France, Switzerland, and Austria. It is presumed that the 
reader is not without an appreciation of the theory of environ- 
mentalism and that certain technical aspects of the character of 
environmental conditions are understood. Where those technical 
aspects are peculiar to mountains, as in the case of temperature 
inversions and Foehn winds, they are explained herein. More- 
over, not all technicalities of mountain environments are in- 
cluded. The physiography and climatology here discussed are 
geographical physiography and geographical climatology. In 
other words, only those facts from physical geography are se- 
lected which are pertinent to a knowledge of human relations. 

This volume is also a bibliographical study. The bibliogra- 
phies are inserted at the close of the several chapters, in order 
that they may be in intimate connection with the discussions 
which the references elaborate. Only occasionally is a title re- 
peated, although some volumes are so comprehensive in scope as 
to deserve a place in several of the reference lists. Ordinarily the 
title is placed in the list which it serves best. Some titles of rela- 
tively slight consequence are included, usually in order to show 
authority for some item of the text. The references lists are 



PREFACE ix 

selected study lists; selected for variety of points of view or of 
location. Not all books studied have been listed; not all books 
listed have been thoroughly studied. Generally I have indicated 
where the data are from nay personal experience. The rest are 
gathered and coordinated from a wide field. Acknowledgments 
for every fact would be so numerous as to annoy the reader. 
Annotations in the bibliographical notes indicate the more im- 
portant sources of data. 

Lastly, acknowledgment should be made to my colleagues for 
helpful criticisms and to Ohio State University for financial 
assistance and leave of absence. 

Roderick Peattie 



CONTENTS 

INTRODUCTION 3 

What is a Mountain? 3 

Attitudes towards Mountains 4 

The Cult of Mountains 6 

I. MOUNTAIN TEMPERATURES . . 9 

Air Pressures 9 

Insolation 11 

Air Temperatures and Altitude 14 

Mass of Mountain 18 

Inversion of Temperature 19 

Growing Seasons and Altitude 23 

Temperature Ranges 25 

Soil Temperatures 27 

II. HUMIDITY AND PRECIPITATION 36 

Humidity and Evaporation 36 

Precipitation 38 

Snow 43 

Human Responses to Snow 50 

Snow Slides 55 

III. WINDS, CLOUDS, AND SUN 63 

Mountains and Winds 63 

Clouds in Mountains 65 

Local Winds 66 

The Foehn Wind 72 

IV. VEGETATION ZONES AND THE HEIGHT LIMITS OF 

FIELDS 79 

Mountain Zones 79 

Height Limits of Fields 85 

The Sunlight Factor 88 

The Geologic Factor 93 

Zones in Andorra 95 

In the Conflent 96 

The Valley of the Doran 98 

V. FORESTS AND THEIR SIGNIFICANCE 107 

Tree Line and Forest Line 107 

Physical Factors 1 1 o 

Plant Factors 115 

Economic Factors 117 



xii CONTENTS 

VI. ALP PASTURES AND ALP ECONOMY 125 

What is an Alp? 125 

Alp Economy 130 

Mountain Nomadism 134 

In the Val D'Anniviers 139 

Pastoral Buildings of the Central Pyrenees 142 

Transhumance: The French Alps 144 

In the Balkans 146 

VII. LAND UTILIZATION AND ECONOMICS 152 

The Economic Balance in Land Utilization 152 

Modern Changes in Land Utilization 158 

Level Land and Terraces 162 

The Factor of Relief: Andorra 165 

VIII. MOUNTAIN POPULATIONS AND THEIR DISTRI- 

BUTION 174 

General Aspects 1 74 

Population and Altitude 177 

Details of Population 181 

Economic Factors 184 

Location of Dwellings 187 

Towns and Villages 189 

Altitude and Health 191 

IX. MATTERS POLITICAL 203 

Mountain States 203 

Mountains as Boundaries 205 

Straddle Economies of the High Pyrenees 210 

An Altitud in al Provincialism: The Tirol 211 

Importance of the Dalmatian Alps 213 

Political Philosophies of Mountaineers 215 

X. THE CHARACTER OF MOUNTAIN LIFE 221 

Mountain Conservatism: Appalachia 221 

Physical Factors in Standards of Living 225 

Seasons in the Alps 228 

The Tenor of Andorran Life 232 

APPENDIX A 

The Metric System 243 

APPENDIX B 

Bibliographical Notes on the Geomorphology of Mountains . 245 



ILLUSTRATIONS 



CHAPTER I 

i Sunlight Circle 13 

2 Hours of Sunlight at Valley Stations 15 

3 Pic du Midi de Bigorre, French Pyrenees 16 

4 A Village in the Sun, French Alps 16 

5 Daily Ranges of Temperature and Altitude 25 

chapter 11 

6 Topography and Precipitation, American Sierra Nevada ... 39 

7 Percentage of Total Precipitation in Snow, Swiss Alps ... 46 

8 Snow Collecting Fields of the Aletsch Glacier, Bernese Ober- 

land 48 

9 Tunnel in an Avalanche, Swiss Alps 48 

10 Snow Limits in the Ortler Alps, Austria 48 

1 1 Snow Free Period in the Oisans According to Altitude ... 52 

12 Piers to Protect Poles against Avalanches, Engadine .... 56 

13 Fences used to Prevent Avalanches 56 

chapter iii 

14 Wind Roses for Switzerland 65 

1 5 Clouds in the Col de Tourmalet, French Pyrenees 68 

16 Clouds after a Snowfall, Hohen Tauren, Austria 68 

1 7 Indraft Winds in the Tarentaise, French Alps 69 

18 Emerging from the Clouds, Spanish Sierra Nevada 72 

19 Valley Clouds in Winter, Karawanken Alps, Austria .... 72 

20 Facts of the Foehn Wind 74 

21 Mechanics of the Foehn Wind 75 

chapter iv 

22 Simple Zoning in the Valley of Llosa, Spanish Pyrenees ... 80 

23 Reversal of Vegetation Zones 83 

24 Sunny and Shady Slope in the Tarentaise, French Alps ... 88 

25 Agricultural Limits, Grisons, Switzerland 88 

26 Sunlight Hours in the Valley of Barcelonnette, French Alps . 90 

27 Field Limits in the Vallee D'Ouell, French Pyrenees 92 

28 High Fields on the Sunny Slopes, Val DTsere, French Alps . . 92 

29 Limits of Culture in the Valley of the Doron, French Alps . . 99 

chapter v 

30 Factors in Tree Limits in the Val Roseg, Switzerland .... 109 

3 1 Tree Limit and Mass of Mountain in Austria 1 1 1 

32 Forest Limit and Mountain Mass in Switzerland 112 



xiv ILLUSTRATIONS 

33 Avalanche Track in the Grisons, Switzerland 114 

34 Slope Farm in the Forest Zone, Stubaital, Austria 114 

35 Example of Forest Distribution in the French Alps 118 

chapter vi 

36 Topographic Relations of Alp Pastures 126 

37 Ground Plan of Up-Mountain Movements, Murau Region of 

Austria 131 

38 Ground Plan of Down-Mountain Movements, Murau Region of 

Austria 132 

39 The Day of the Cattle Down-Drive in Andorra 134 

40 Transhumance in Andorra 134 

41 Mountain Nomadism in Three Regions of the French Alps . . . 136 

42 Seasonal Movements in the Val D'Anniviers, Switzerland . . 140 

43 Pastoral Establishments near Pralognon, French Alps ... 142 

44 Cheese Huts of the Marjelen Alps, Bernese Oberland, Switzer- 

land 142 

chapter vii 

45 Land Use in the Val D'Isere, French Alps 153 

46 Chorography of the Conflent, French Pyrenees I 156 

47 Chorography of the Conflent, French Pyrenees, II .... 157 

48 Cultivation of a Forty-Five Degree Slope, French Pyrenees . 160 

49 Terraced Farm in Provence, France 160 

50 Relation between Slope and Culture, Central French Pyrenees 166 

51 Land Use in a Canyon, Andorra 166 

52 Land Utilization Map of Andorra 168 

chapter viii 

53 Population by Altitude in the Oisans, French Alps 1 78 

54 Permanent House Limits and Grain and Forest Limits in the 

Ortler Alps, Austria 180 

55 Population about DOrfli, Switzerland 186 

56 Winter Village in the Valley of the Plessur, Grisons .... 194 

57 Summer Village in the Valley of the Sapunerbach, Grisons . 194 

chapter IX 

58 Boundaries of Andorra 206 

59 Hay Transportation in the Pyrenees 210 

60 Loggers in Andorra 210 

61 Mounting to the Potato Fields, Spanish Sierra Nevada ... 216 

62 Bringing Rye to the Village, French Pyrenees 216 

chapter x 

63 The Road to Mantet, French Pyrenees 226 

64 Road Engineering, Switzerland 226 

65 Transportation in the Albula Pass Region, Switzerland . . . 230 



INTRODUCTION 

What Is a Mountain? 

A MOUNTAIN, strictly speaking, is a conspicuous elevation 
of small summit area. A plateau is a similar elevation of 
larger summit area with at least one sheer side. An essential and 
yet indefinite element in the definition of a mountain is the con- 
spicuity. Conspicuity, like height, is a relative matter, and de- 
pends upon the personal evaluation or the standard by which it is 
measured. Many eminences but a few hundred feet high are 
termed mountains by dwellers on flat plains. One writer arbi- 
trarily states that a mountain must be a quarter of a mile high. 
If this relief be measured from the surrounding country rather 
than from sea level, then certainly one would have a mountain. 
Seldom is relief as great as on the coast of Formosa, where there 
is a precipitous cliff of 4270 meters. 1 The Great Plains of the 
Western United States are a mile high. A slight eminence upon 
these plains would hardly be termed a mountain. Pikes Peak is, 
in truth, a mountain not because it rises more than 4270 meters 
but because its relief over the surrounding country is so great 
(2440 meters). Also it has steep sides. Its conspicuity is great. 
For days, in the era of traveling by ox cart, its white summit was 
a guide to the early settlers, who bore upon their covered wagons 
the slogan " Pike's Peak or Bust." It was a symbol, a goal, and it 
played a great part in the imagination of the plodding, hopeful 
travelers. Mountains should be impressive; they should enter 
into the imagination of the people who live within their shadows. 
Unfortunately it is next to impossible to include such intangibles 
in a definition. Mountains have bulk; mountains have also 
individuality. 

The element of ^ndjyi dualit yl- is not a far-fetched phrase. 
Fujiyama and Mount Etna are mountains of the same type, and 
yet they have individual characters. They are isolated volcanic 
cones, steep near the summit, with gentler curves in their lower 
reaches. Both are snow-capped. Both are majestic watchers of 
the human activity which mills about their bases like the con- 

1 For conversions to and from the metric system see Appendix A. 



4 MOUNTAIN GEOGRAPHY 

fusion of ant hills. Alike geologically and topographically they 
have different psychological reactions in the minds of people who 
daily regard them. Fuji is benign. Its serenity gives it a place in 
Japanese philosophy. It is sacred and it is the most common 
motive of Japanese art. Etna, if one is a dualist, is a devil rather 
than a divinity. It is a force for evil, whose boiling arms of lava 
reach out fiendishly towards the villages. 

Attitudes Towards Mountains 

To a large extent, then, a mountain is a mountain because of 
the part it plays in popular imagination. It may be hardly more 
than a hill ; but if it has distinct individuality, or plays a more or 
less symbolic r61e to the people, it is likely to be rated a mountain 
by those who live about its base. In the days of the Greeks, 
mountains such as Parnassus (2458 meters) and Olympus (2972 
meters) were mysterious regions where dwelt the gods, and in 
whose surrounding glens roamed satyrs and oreads. Possibly 
because of their detached splendor, mountains were sacred in 
early days, and the tradition is carried over into modern times. 
Among the sacred mountains are Ararat in Armenia and Leb- 
anon in Syria. China has at least five sacred peaks. One of 
these, Omei, a peak in Szechuen, rising to a height of 3098 
meters, has fifty-six pagodas and thirty-five monasteries and 
temples for those faithful to Buddha. It was from a mountain 
that Buddha ascended into heaven. It was on Mount Sinai that 
Moses received the laws. David decided upon Mount Zion as 
a site for his capital; Abraham took Isaac to a mountain in the 
land of Moriah to sacrifice him to Jehovah; and throughout the 
songs of the Israelites there runs a current of veneration for 
mountains. 

In the Middle Ages a certain dread of mountains was evident. 
Dante makes mountains the guardians of the gates of Hell. The 
Scandinavians peopled them with gnomes who were vassals of 
the Ice Queen. On Walpurgis Night all the witches of the earth 
and air danced in the Harz Mountains of Germany. In moun- 
tains were supposed to dwell those mystic folk, myrmidons, 
pygmies, fingerlings, fairies, and specters. So great was the fear 
inspired by mountains and wild gorges that in 1401, when Adam 
of Usk went on a pilgrimage to Rome, he was carried blindfolded 
over the St. Gotthard Pass in order that he might not rest his 



INTRODUCTION 5 

eyes upon the fearfulness of the scene. Many an overlord passing 
through the mountains secured himself by a large military escort. 
Such was the fear inspired by the rugged heights that travelers 
wrote back in report upon the passage to say that they were safe 
in body and in soul. 

In 1 5 1 1 , on the other hand, Luther on pilgrimage to Rome 
spoke of the pleasant life "of Switzerland," and said that the 
miles of that country were the "shortest miles." Yet Benvenuto 
Cellini looked in 1537 upon the Alps as full of dangers. In mak- 
ing the passage he was accompanied by a cavalcade and wore a 
shirt of mail. It was in the eighteenth century that the moun- 
tains came to be loved for themselves. A scientist and poet of 
Zurich, Albrecht von Haller, in 1732 wrote a long poem, Die 
Alpen, which did much to make known the beauty of the moun- 
tain regions. It is said to have been Swiss out-of-door literature 
that incited Rousseau passionately to preach his return to na- 
ture. Horace Benedicte de Saussure, the first great Swiss Alpinist 
and an admirer of Rousseau, attempted the ascent of Mont 
Blanc four times between 1 760 and 1 787. This was a stupendous 
task in terms of the times. Saussure was so little acquainted with 
the problems of the first ascent that he took with him a sunshade 
and smelling salts. Veils were worn against snow-blindness. A 
Jacques Balmat was the first to reach the summit. For this feat he 
received the title ' Mont Blanc 5 from the king of Sardinia. Saus- 
sure in 1 787 was the first to follow him. 

It was then that the English discovered the Alps. They were 
the first tourists in numbers to visit the out-of-the-way valleys 
and climb the peaks. True, the Tillis in 1791, the Jungfrau in 
181 1, the Finsteraarhorn in 181 2, and the Schreckhorn in 1842 
were conquered by Swiss. That Hudson, Hadlow, and Lord 
Douglas lost their lives on Mont Cervin (Matterhorn) in 1 865 was 
sufficient to challenge the English. From then on they literally 
flooded the remote valleys of Switzerland. The Swiss scientific 
interest gave way to the English sense of sport. The Alps became 
a subject for prose and poetry. One should not fail to mention 
the public imagination as reflected in two great poems. Schil- 
ler's poem, Wilhelm Tell, did for Central Switzerland what 
Byron's Childe Harold did for French Switzerland. 

There are two attitudes which men today hold towards moun- 
tains. One is the attitude of the mountain climber and the other 
is that of the scientific man, whether he be geologist, geographer, 



6 MOUNTAIN GEOGRAPHY 

or climatologist. Nor indeed are the two attitudes always dis- 
tinct. 

The Cult of Mountains 

There has grown up among travelers and sportsmen a cult of 
mountains which is a modern and more conscious phase of the 
old worship of mountains. Particularly have the English, Ger- 
mans, Austrians, Swiss, Italians, and French evidenced this. 
Each nation has its important Alpine club and journal. The 
journals, concerning themselves in part with the technique of 
mountain climbing and routes of ascent, also are devoted to 
praise of mountains. The best of the writings of Englishmen on 
mountains has been collected by Arnold Lunn in one volume, 
from which the following quotations are taken. 

Shelley in his History of a Six Weeks 9 Tour expresses himself on 
Mont Blanc thus: 

Mont Blanc was before us — the Alps, with their innumerable gla- 
ciers on high all around, closing in the complicated windings of the 
single vale — forests inexpressibly beautiful, but majestic in their 
beauty — intermingled beech and pine, and oak, overshadowed our 
road, or receded, whilst lawns of such verdure as I have never seen be- 
fore occupied these openings, and gradually became darker in their 
recesses. Mont Blanc was before us, but it was covered with cloud; its 
base, furrowed with dreadful gaps, was seen above. Pinnacles of snow 
intolerably bright, part of the chain connected with Mont Blanc, shone 
through the clouds at intervals on high. I never knew — I never 
imagined what mountains were before. 

Edward Whymper, famous among Alpinists, in his Scrambles 
among the Alps in the Tears iS6o—i86g y describes the view from a 
lonely bivouac on the Matterhorn: 

I returned to the view. The sun was setting, and its rosy rays, blend- 
ing with the snowy blue, had thrown a pale, pure violet far as the eye 
could see; the valleys were drowned in purple gloom, whilst the sum- 
mits shone with unnatural brightness; and as I sat in the door of the 
tent, and watched the twilight change to darkness, the earth seemed to 
become less earthy and almost sublime; the world seemed dead, and I, 
its sole inhabitant. 

Douglas Freshfield, another famous Alpinist, in The Italian 
Alps, finds hardly sufficient adjectives to express the beauty about 
him. 



INTRODUCTION 7 

The full midday glow of a July sun was falling from the dark va- 
pourless vault overhead on to the topmost crags of Monte Rosa. A 
delicate breeze, or rather air-ripple, lapping softly round the mountain- 
crest, scarcely tempered the scorching force with which the rays fell 
through the thin atmosphere. Round us on three sides the thousand- 
crested Alps swept in a vast semicircle of snow and ice, clustering in 
bright companies or ranging their snowy heads in sun-tipped lines 
against the horizon. 

F. W. Bourdillon in the Alpine Journal (vol. xxiv) speaks of the 
love of mountains. 

I suppose this ideal love of mountains — this love that we may almost 
call a platonic love, since it seeks no selfish gain — really exists in most 
or all of us; and is at the root of the instinct certainly of the climber, pos- 
sibly even of the tourist. We have all of us had our c moments,' either 
on the mountains, or perhaps in some distant view of them, when life 
and joy have assumed new meanings, and the world's horizons sud- 
denly broken down and shown us realms of dream beyond and yet be- 
yond. Sometimes it is on the top of some lonely peak, when the world 
seems at our feet, and the blue dome of space an appreciable thing; 
sometimes it is among the hush of snow-fields and glacier-walls, with 
icy peaks above and moonlit mists below us; sometimes it is from some 
lower height, where suddenly a panorama of silver tops breaks on us, 
or we see the far-distant snow peaks mirrored in sunny lake waters. 

Other than the visitor to mountains there is the dweller among 
the peaks who, familiar with the masses about him to the point 
of almost personal friendship, finds in the majesty of alpine pin- 
nacles or the dearness of lesser hills, their ever-changing aspect in 
light and shade, calm and storm. 

Hardly less strong is the scientific interest in the multi-fold 
phases of mountains and mountain life. A bibliography of the 
scientific works on mountains would be overwhelming. The 
store of knowledge which has resulted from the studies would 
itself be of mountainous proportions. It is a brief of this many- 
sided research that this book proposes to undertake. Incom- 
plete as it must be, to be contained within the covers of a single 
volume, it is hoped that it will yet present in resume sufficient 
material to enable the reader to understand the general factors 
in the human geography of mountains. 



8 MOUNTAIN GEOGRAPHY 

BIBLIOGRAPHICAL NOTES 

Human Attitudes Towards Mountains 

Coolidge, W. A. B. Alpine Studies. London, 191 2. 

Coolidge, W. A. B. The Alps in Mature and History. New York, 1908. 

De Beer, G. R. Alps and Men. London, 1932. 

Engel, Claire E., and Vallot, Charles. Les ecrivains a la montagne: "Ces 

monts qffreux" (1650-1816). Paris, 1934. 
Fay, C. E. "The Mountain as an Influence in Modern Life," in Appa- 

lachia, xi (1905), pp. 27-40. 
Godley, A. D. "Mountains and the Public," in Alpine Journal , xxxvii 

(1925), pp. 107-117. 
Hamerton, P. G. Landscape. London, 1885. 
Hedin, Sven. Southern Tibet. Stockholm, 1916-22. 9 vols. Vol. vii, 

pp. 9-10. 
Hyde, W. W. "The Ancient Appreciation of Mountain Scenery," in 

Classical Journal, xi (191 5-1 6), pp. 70-84. 
Hyde, W. W. "The Development of the Appreciation of Mountain 

Scenery in Modern Times," in Geographical Review, iii (191 7), pp. 

1 07-1 18. These two articles by Hyde are very valuable. 
Lunn, Arnold. The Englishman in the Alps, 2d ed. London, 1927. A 

delightful compendium of literary extracts. 
Perry, T. S. " Mountains in Literature," in Atlantic Monthly, xliv (1879), 

pp. 302-311. 
Reclus, Elisee. The History of a Mountain, tr. from the French by Bertha 

Ness and John Lillie. London, 1881. Chaps, xvii, xix, xx, xxi, xxii. 
Stutfield, H. E. M. "Mountaineering as a Religion," in The Alpine 

Journal, xxxii (19 18), pp. 241-247. 
Tozer, H. F. A History of Ancient Geography. Cambridge, England, 1897. 

Chap. xv. 
Van Dyke, J. C. The Mountain. New York, 191 6. 



CHAPTER I 
MOUNTAIN TEMPERATURES 

Air Pressures 

THE climate of any mountain region is, as with lowlands, 
primarily determined by latitude, prevailing winds, and con- 
tinentality. But mountains have, as definite factors in their cli- 
mates, the matters of altitude and exposure. 

Altitude modifies: 

Air pressures. 

Air composition. 

Insolation. 

Air temperatures. 

Temperature ranges. 

Soil temperatures. 

Winds. 

Evaporation. 

Humidity. 

Clouds. 

Precipitation. 

Snow percentages. 

Exposure, by its contrasts, exaggerates or modifies the quan- 
tity of the altitudinal modifications. These contrasts are: 

Sunny slopes and shady slopes. 
Wet slopes and dry slopes. 
Windy slopes and protected slopes. 

How greatly these modifications are influential in the lives 
and economies of men it is hoped this volume will point out. Of 
necessity the following discussion treats the various factors and 
conditions separately. It is their interrelation, of course, that 
makes up the component whole. A warning is then made not to 
forget that it is the total character of the climate which is the 
significant concern. We are not here concerned to present a com- 
plete discussion of climatology. Only those aspects which are of 
human importance are discussed, and the treatment is thus truly 
geographic. 

One introductory matter which must be thoroughly under- 



IO MOUNTAIN GEOGRAPHY 

stood is the altitudinal decrease of pressure. If the pressure of the 
air at sea level at 20 degrees centigrade is normally 762 milli- 
meters, the following decrease, other things being equal, will 
exist: 

Pressure and Altitude 





Mean pressure in 
millimeters 


Change in 
each mm 


altitude in meters for 


Altitude in meters 


1. of pressure change 


O 


762 




IO.5 


5OO 


7*9 




I I.I 


IOOO 


678 




11.8 


15OO 


639 




I2 -5 


2000 


6oi 




13-4 


25OO 


566 




14.2 


3OOO 


532 




!5-i 


3500 


5OI 




16.1 


4OOO 


470 




17.2 


5OOO 


415 




19.6 


6000 


3 6 4 




22.5 



Pressure in millimeters represents dead weight. Millibars represent energy. To 
convert the weight symbol to the energy symbol multiply by 1.35. 

It will be seen that the rate of decrease of pressure with alti- 
tude is not regular. Following is a formula for computing the 
rate of decrease. If pressure at sea level is 1 003 kilograms for one 
cubic centimeter, or 762 millimeters, the decrease with altitude 
is at first 1 millimeter for every ten meters, but at higher levels 
the rate of decrease is i .9 millimeters. For further pressure reduc- 
tion formulae, one should consult Knoch's edition of Hann. The 
above table is general, but serves the purpose of all save the most 
exacting. A rule of thumb is that one-tenth of 1 millimeter of 
pressure decrease approximates 10.5 meters of altitude change. 
One inch of pressure equals about 90 meters. This rule is not 
accurate above 900 meters or 3000 feet. Moreover, the pressure 
decrease varies with weather, latitude, and variation of vertical 
temperature gradients. Yet the decrease of pressure with alti- 
tude is the most regular of all mountain climatological phe- 
nomena. 1 

1 A good aneroid barometer with a movable disc to indicate altitudes for climbers 
and surveyors is properly graduated to compensate for this differing rate of decrease. 
An altimeter is merely an aneroid barometer so equipped with a movable scale that 
as decrease of pressure is brought about by ascent the same needle indicates the in- 
crease in elevation. Interestingly enough, an early test of the barometer was ac- 
complished by scaling a mountain. Blaise Pascal in 1648, five years after the 
invention of the barometer, persuaded his brother to climb Puy-de-D6me in the 
Massif Central of France. Simultaneous observations were made at the base and 
the top of the mountain. The aneroid barometer was invented by Vidi in 1 848. 



MOUNTAIN TEMPERATURES 1 1 

Following are some mean pressures of high level stations. 

Pressure for High Meteorological Stations and 
Observation Points 

Station Latitude Height Average pressure 

Quito o° 14/ S 2850 m. 547-5 mm. 

Bogota, Colombia 4° 35' N 2660 558.4 

Mexico 19 2 6' 2278 586.3 

Etna, Sicily 37 44' 2950 534.4 

Pic du Midi, France 42 57' 2859 539.5 

Mont Blanc, France 45 50' 4359 447-° 

Sonnblick, Switzerland 47 3' 3 I0 5 5*9-7 

Kloster Hanle, Tibet 22 40' 4610 435-4 

Decreasing pressures have significance in many aspects of 
climatology. They determine the inherent or dynamic heat of 
gases, they affect the dew point, and, of great human concern, 
they have distinct physiological effects upon the body. Appre- 
ciating, then, the decrease of air densities with altitude, we are 
prepared to take up the matter of insolation, that is, the action or 
effect of the sun's rays on a body exposed to them, and the conse- 
quent temperatures. 

Insolation 

There are three conditions affecting insolation received by 
mountains, which are purely results of local topographic modi- 
fications. 1 

Altitude. 

Angle of exposed slope. 

Position in the local relief. (Figures 1 and 2.) 

The percentage of insolation received by the earth increases 
with altitude. This has two causes. 

One cause is the density of air. On lower levels smaller 
amounts of insolation are received because the subtraction of 
radiant energy by the absorbent air is great. At 2450 meters one 
is above one-fourth of the atmosphere by weight, and at 5800 
meters above one-half the atmosphere by weight. 

1 Too late for incorporation in this work is the extremely interesting and thor- 
oughly scientific discussion of insolation by Alice Garnett, " Insolation, Topography, 
and Settlement in the Alps," in The Geographical Review, xxv (1935), pp. 601-617. 
This is well worth the reader's attention. It represents an advance in the graphic 
representation of insolation data. 



12 MOUNTAIN GEOGRAPHY 

The lower layers of the atmosphere have high absorbent quali- 
ties, not only because of actual mass of material but also because 
of the quality of the material. Water vapor, carbon dioxide, and 
dust all absorb greater quantities of heat than other elements of 
the atmosphere. All are chiefly found near the earth. At 2450 
meters one is above one-half the atmospheric moisture and more 
than one-half the suspended dust. The heavy carbon dioxide in 
calm air clings to the earth. Above 900 meters liquid and solid 
impurities have slight influence on the amount of insolation re- 
ceived. An exception is found in the humidity present during 
temperature inversions. This freedom of the atmosphere from 
impurities at high levels is indicated by the brilliancy of sunlight 
as well as the actual sighting of stars after sunrise. Thus Orion 
has been seen after sunrise from the Jungfraujoch in the Bernese 
Oberland of Switzerland, at an elevation of 3454 meters. 

On clear days a rock surface at sea level receives 50 per cent of 
the possible insolation. At 1800 meters altitude some 75 per 
cent of the possible insolation reaches the rock. 

Percentage of Insolation Received at Certain Altitudes 

Station Altitude Per cent of Insolation 

Mont Blanc 4807 m. 94 

Grands-Mulets 3050 80 

Bossom glacier 1200 79 

Grenoble 215 71 

At Leh (3500 meters in the Himalayas) water was boiled by the 
sun when it was exposed in a blackened dish set in a transparent 
bottle. The difference between temperature in a black bulb 
thermometer in a vacuum and an ordinary thermometer in- 
creases with increasing altitude. 

Not only is there more insolation at high levels but also a dif- 
ferent quality. The ultra-violet rays are there more active and 
the chemical action of sunlight increases with altitude. The dif- 
ference between the normal air temperature and the actinic 
temperatures accounts for the high sensible temperatures of high 
altitude. In formula, the sensible temperature equals 12 xl, 
where / equals the sun's rays in gram-calories per cubic centi- 
meter per minute. The multiplier may be increased to 20 when 
the snow serves as a reflector. At Davos / has equaled 1 .46 gram- 
calories. 

12 (1.46) = 17.5 degrees Centigrade. 
20 (1.46) = 29.2 degrees Centigrade. 



MOUNTAIN TEMPERATURES 



13 




Fig. 



Sunlight Circle 



The point of observation is the center of the circle. The top and bottom arcs 
represent, each, the summer and winter solstice. The middle arc is the path of the 
sun at the equinoxes. The shaded portion is the horizon of surrounding mountains 
as seen from the point of observation, which in this case is a station in the valley of 
the Veneon, French Alps. (After Allix.) 

Comfort has been found above a snow cover at Davos when air 
temperatures were as low as —10 degrees. As if by way of com- 
pensation, high valleys suffer less from low air temperatures be- 
cause of high insolation values. The high insolation means that 
where direct sunshine is found there are high soil temperatures 
and high plant temperatures. Therefore an increase of altitude 
is not exactly coincident with a decrease in economic possibili- 
ties. Plant zones are higher on mountain slopes than would be 
imagined from considering air temperatures. Heat in high alti- 
tudes is a response to insolation rather than to air temperature. 



14 MOUNTAIN GEOGRAPHY 

Relations between air temperatures and insolation values are 
suggested by the following measurements made by the writer. 
They were observed in a cirque (2250 meters) near the Porte de 
Vennasque, Central Pyrenees, at ten in the morning of a July 
day. Whereas the actinometer read 46.5 degrees, the air tem- 
perature in the shade was but 13 degrees. The writer also made 
observations of the sun's heat at Langweis in the Grisons x at 
1383 meters. The shadow of the mountain in the afternoon of 
October first rapidly passes over the town at about nine minutes 
to four. Measurements were taken with the thermometer in the 
sun at nine minutes to four, again at four minutes to four and 
lastly at one minute after four, that is, at five minute intervals. 

Temperature just before sundown: 27. 5 C. 
Temperature five minutes later: 21.25 G. 
Temperature five minutes later: 17.5 G. 

This illustrates how in rare air the temperature is largely insola- 
tion. The air retains little heat once the sun is down. 

Hann (ed. by Knoch) shows the increasing difference between 
temperatures in sun and shade with increasing altitude. 

Temperatures in Sun and Shade 

Difference of temperatures 
Place Elevation in sun and shade 

Riffelberg 2570 m. 2 1 .o° C. 

Hornle 2890 m. 28.0 

Gornergrat 3140 m. 32. 8° 

Air Temperatures and Altitude 

Generally speaking, air temperatures decrease with altitude. 
Exceptions to this generality exist and will be discussed later. 
The rate of decrease depends upon the composition of the air, the 
degree and character of the slope, the direction of the exposure, 
the mass of the elevation, the vegetal covering, and the existing 
wind currents. The fundamental reasons for decrease are: 

Increasing rarity of air with increased altitude signifies actually 
less molecular material to receive and hold the heat. The table 
showing air pressures is a measure of this rarity. 

1 Grisons is the French and English, Graubunden the German name of the largest 
and easternmost canton of Switzerland. I have used the two names indiscrim- 
inately. 



MOUNTAIN TEMPERATURES 



15 




Fig. 2. Hours of Sunlight at Valley Stations 

For ten stations on the sunny slope and five stations on the shady slope in the 
Valgaudemar, French Alps. The hours between the black areas represent the 
length of sunlight on the longest day of the year and the unshaded areas the hours 
on the shortest day. (Courtesy of the Geographical Review, published by the American 
Geographical Society of New York.) 



A decrease with altitude of two principal heat-absorbing gases 
which are common 'impurities' of low air levels. These are car- 
bon dioxide and water vapor. 

Increasing rarity of air means less dynamic molecular heat. 
This decrease of dynamic heat is the adiabatic cooling. 

The rate of adiabatic cooling is 1 degree centigrade for every 
100 meters, or 1.6 degrees Fahrenheit for every 300 feet. De- 
creasing the pressure of a gas is the equivalent of permitting 



1 6 MOUNTAIN GEOGRAPHY 

fewer molecules to occupy a given space. The number of im- 
pacts of molecules within the space will then be fewer and con- 
sequently the heat resulting from the impacts of the molecules 
will be less. We shall speak of this molecular heat of a gas as the 
inherent heat. 

The lower levels of the atmosphere absorb much more heat 
than do the equivalent upper layers of the atmosphere. The 
amount of heat absorbed by a layer of air depends upon the 
molecular density and the composition. The chief elements, or, 
better, impurities of the composition important in this connec- 
tion are dust, carbon dioxide, and water vapor. The air is not so 
constituted as to absorb easily the short wave length radiation of 
the sun. Air temperatures are raised much more largely by the 
long wave lengths of earth radiation and by conduction. 

The actual rate at which temperature at any instant may be 
found to decrease aloft is the vertical temperature gradient. 
This gradient of actual temperature is different from the adia- 
batic decrease, since a number of other factors enter into the mat- 
ter. The first measurement of the vertical temperature gradient 
in a mountainous region was made in 1788 by Saussure in the 
Alps. He found 1.6 degrees for 88.76 meters. The mean for 17 
extra-tropical mountains has been computed as 0.57 degrees for 
100 meters. Pikes Peak, Colorado, in a dry atmosphere gave 
0.63 degrees for 100 meters. Shreve in the Santa Catalina Moun- 
tains, Arizona, has found the gradient as high as 0.76 degrees for 
102 meters. Friih in a recent work states that the gradient for a 
mean of 169 Swiss stations is 0.52 degrees. Hann in his maps 
used 1 degree for every 200 meters. Bartholomew in his classic 
Atlas of Meteorology used the scale of 1 degree for every 150 
meters. Thus they both reduced temperatures of the world to 
sea level temperatures. The advantage of such a map is that, 
confusing details being eliminated, one is able to see the general 
controls of climate. The mountain geographer wishes to have 
actual temperatures of the high stations. He may obtain these 
by dividing the height of the station in meters by 150, or 200 as 
the case may be, and subtracting the result from the surface 
temperature. 

Alfred de Quervain has written upon methods of studying the 
vertical temperature gradient. He shows the influence of such 
factors as mass of mountain, cloudiness, exposure, and season. 
A. J. Henry has five articles of significance not only in presenting 




Fig. 3. Pic r>u Midi de Bigorre, French Pyrenees 
This isolated peak has the highest permanently inhabited house in Europe. It is a 
meteorological station with all-year observations. The foreground is a typical 
summer pasture for cattle. The higher slopes are available for sheep. 




-fevl 



- * :■'■ 

m - • 1 1 - 
- ■*-• • -- • |9b 



Fig. 4. A Village in the Sun, French Alps 
This village and its fields lie entirely on the sunny slope of a valley. The stream 
marks the edge of the forest on the shady slope. The scene is taken in one of the 
hanging valleys of the Tarentaise. (Courtesy of Ginn and Company, New College 
Geography. ) 



Spring 


Summer 


Autumn 


Year 


O.60 


O.62 


O.47 


O.51 


O.66 


O.67 


0.57 


O.60 



MOUNTAIN TEMPERATURES 17 

actual data on the vertical temperature gradient, but also in 
illustrating the importance in weather forecasting. Moreover, 
the studies especially demonstrate the importance of exposure. 

The mean for Switzerland is sufficiently accurate so that the 
mean temperature of a station may be computed approximately 
from its altitude. Thus Zurich at 493 meters elevation has a 
mean temperature of 8.5 degrees. Uto Kulm is in elevation 389 
meters higher than Zurich. The temperature of Uto Kulm has 
by this formula a mean temperature of 6.5 degrees; 8.5-3.8(.5i). 
Maurer uses this mathematical procedure for finding tempera- 
tures of alpine stations. 

The seasonal differences in vertical temperature gradients have 
been observed by many. De Martonne gives the following table 
to illustrate differences of gradients according to season and to 
exposure. 

Temperature Gradients in the Alps 

Winter 

North slope 0.35 

South slope 0.50 

From this it will be seen that the most rapid gradient is in the 
spring, when the upper slopes are covered with snow. The south- 
facing slope, in this hemisphere, has the more rapid gradient 
because of the greater warming at the base as compared with the 
north slope. 

What are the actual temperatures on mountains? If we con- 
sider the mean annual temperatures we find that similar tem- 
peratures are found at lower altitudes as the latitude increases. 

Elevation of Mean Annual Isotherm of Zero Degrees 

Place Elevation in meters 

Andes, near Quito 5 1 00 

Northwest Himalayas 4700 

Mount Etna, Sicily 2 95<> 

Pic du Midi, French Pyrenees 2480 

Swiss Alps 2200 

Ben Nevis, Scotland 1250 

Though a variety of factors must be considered in the analysis 
of the climate of any one station, latitude, continentality, and 
mass of mountain stand foremost. The latitude control is illus- 
trated by the above table. Continentality suggests irregularity in 
the data for stations of the same latitude. Thus Pikes Peak, Colo- 



1 8 MOUNTAIN GEOGRAPHY 

rado, which is not far from the latitude of Mount Etna, has an 
elevation of the mean annual zero degree isotherm of 3200 me- 
ters, as compared with 2950 meters for the Sicilian station. 



Mass of Mountain 

The influence of mass of mountain (Massenerhebung) is indicated 
by the extraordinary fact that in the Himalayas at 4000 meters it 
is seldom below freezing from the end of May to the middle of 
October. Yet among the peaks at 5000 meters, where the mass is 
considerably less, it seldom is above that point even in summer. 
Other things being equal, the more massive the bulk of a moun- 
tain the greater the mean temperature at a given elevation. 
More significant is the fact that the greater the mass, normally 
the longer is the growing season. The most casual traveler among 
mountains cannot fail to notice the higher limits of agriculture on 
plateaus than on peaked mountains. Thus rye culture is nota- 
bly higher on the plateau of the Cerdagne in the French Pyre- 
nees than on the flanks of the mountains of the neighboring 
Conflent. Topography and accessibility might be thought to be 
hindrances to rye growing in the latter region, were it not that 
rye is grown in tiny fields in almost inaccessible spots. Forest 
limits are higher in the massive Alps than in the Fore-Alps of 
France. Imhof points out the same relationship for Swiss forests. 
Marek puts this fact in map form. Mass of mountain can be 
worked out quantitatively. Anfossi gives a formula for its com- 
putation. 

Culture lines, or lines of human occupation and use, are higher 
on mountains of great mass because of a number of other reasons. 
Greater mass means ordinarily gentler slope. Roads mount such 
slopes easily. Fields and villages are thus found at higher levels. 
Mass generally increases the cultural heights in the Rhaetian 
Highland and the Central Alps of the Tirol in distinction to the 
limits in the less massive Eastern Alps. Because of greater snow- 
fall, the character of the soil (mountains of great mass are com- 
monly crystallines), and the degree of slope, there may be exces- 
sive ground water which forces the cereal lines valley ward. 
Cereal lines are therefore not in every case higher on plateau 
slopes than on the flanks of peaks. 

The actual rate of decrease of temperatures on a given moun- 
tain range depends upon the adiabatic rate of decrease, the mass 



MOUNTAIN TEMPERATURES 



*9 



of the successive zones, the angle of slope, and the exposure, as 
well as other details of topography, vegetation, and winds. An 
actual example of decreasing temperatures with altitude is the 
following from the Carpathian Mountains: 

Temperature and Altitude in the Carpathian Mountains 

Place Height in meters 

Crakow 220 

Turka 580 

Poronin 742 

Jablonica 900 

Zakopana 1000 

Jawarzyna 10 19 

For the sake of geographic variety let us take certain data from 
Hettncr's study of Colombia. The table is generalized. 



Jan. 


Apr. 


July 


Oct. 


Year 


-34 


8.0 


19.0 


8.3 


7-9 


-5.8 


6.2 


16.7 


7.3 


5-9 


-5-6 


5° 


15-3 


6.1 


4-9 


-8.4 


4.8 


14.7 


6.4 


4.2 


-54 


4-3 


14-3 


5-5 


4-3 


-6.8 


3 1 


13.8 


3-9 


34 



Temperatures and 


Altitudes in the Tropics 


Altitude 


Mean annual temperature 


400 meters 

600 

800 


27.4' 

26.2 

25.O 


Tierra calien 


1000 


23.8 




1200 


22.6 
21.3 
20.2 




1400 
1600 


Tierra tempi 


1800 


I9.O 




2000 


18.O , 




2200 


17. 1 ] 




2400 
2600 


I6.I 


Tierra fria 


2800 


*3-9 




3000 


12.7 ; 




3200 


"•5 




3400 
3600 


10.3 
9.1 


• Paramo 


3800 


8.0 




4000 


7.0 





Inversions of Temperature 

It is not always true that temperatures decrease with altitude. 
For a distance up temperatures may temporarily increase. This 
is known as an inversion of temperature. Thus an inversion of 
temperature exists when, upon ascent, a plus rather than a minus 
temperature gradient of air temperatures is experienced. Above 



20 MOUNTAIN GEOGRAPHY 

this zone the temperature will decline. These warmer zones are 
known as thermal belts. 

Examples of this phenomenon are many. In America the best 
known and well studied observation is on the front of the Blue 
Ridge Mountains of North Carolina. Here over a four-year 
period minimum temperatures averaged lower on the valley 
floors than in the thermal belts. On one slope, where there was 
527 meters difference in elevation between the lowest and highest 
thermograph, the average minimum temperature was 1.84 de- 
grees C. higher on the summit than at the base. The greatest 
inversion observed in this region was a plus gradient of 3 1 degrees 
F. in 1000 feet (17.2 degrees C. in 305 meters). The best known 
European inversion occurs in the Klagenfurt basin in Carinthia. 
Here it is at times necessary to mount 1 000 meters to find again a 
temperature as low as that of the valley floor. In the following 
table the underlined figures represent the limit of the inversion 
according to Machatschek. 

Temperature Inversion in the Klagenhjrt Basin 

Station Altitude Jan. temperatures Winter average 

Klagenfurt 440 m. —6.2 —4.6 

Ebcnstcin 570 —4.2 —3-3 

Huttcnberg 780 — 3.1 —2.3 

Lolling Tal 840 —2.5 —1.6 

Lolling Berghaus 1 100 -~ l -9 "~ I -3 

Stelzing 14.10 — 3-7 ~3- 2 

The Engadinc and the Upper Drave Valley have well known 
inversions. Sils in the Engadinc at 181 1 meters has a January 
temperature of —0.8 degrees, whereas St. Bernhard at 2478 
meters is 8.3 degrees. Bcvers, 75 meters below the summit of the 
Rigi, is 4.6 degrees colder. It is a general rule that isolated peaks 
in winter are warmer than high side valleys. This is demon- 
strated by figures from the Klagenfurt region given by Supan. 

High Valleys and Peaks in the Klagenfurt Basin 

Station Altitude, meters January Year 

Klagenfurt 440 —6.2 7.2 

Kappel 560 —5-2 6.6 

Fellach 805 —4.0 6.0 

Unterschaffler Alps 1063 —3.6 5.5 

Obir I 1230 —4.3 4.7 

Obir II 1612 —5.1 3.7 

Hoch-Obir 2047 —6.8 0.6 



July 


Aug. 


Sept. 


Oct 


Nov. 


Year 


9-9 


9-4 


7-5 


2-7 


-O.8 


2.0 


"•3 


10.4 


7-4 


2.4 


-24 


1-5 


1.4 


1.0 


— O.I 


-0.3 


-1.6 


-0.5 



MOUNTAIN TEMPERATURES 21 

A high valley at even greater altitude than an isolated peak may 
demonstrate this relationship. This is perhaps a demonstration 
of mass of mountain as well as inversion of temperature. Com- 
parative temperatures between a valley station, Sils-Maria, and a 
peak, Rigi, follow. 

Temperature on Sils-Maria and on the Summit of Rigi 

Jan Feb. Mar. Apr. May June 

Ri gi -3-9 -4-5 -3-9 -°- 2 3-o 7-5 

Sils —6.8 —8.0 —4.0 —06 5.1 9.1 

Difference — 2.9 —3.5 —0.6 —0.4 2.1 1.4 

Though inversions are a well known phenomenon of free air, 
inversions in mountain valleys differ because they are much in- 
tensified and stabilized. There have been many students of in- 
versions, both observers and theorists. Benevent is particularly 
lucid in his explanations. Friih has the best descriptions. Atten- 
tion is called to the excellent and complete presentations of 
theories of inversion offered by Napier Shaw. The presentation 
in this present volume is generalized. Geographers are more 
concerned with the fact of inversion than the theories of origin. 

The simplest type of inversion of temperature comes about 
from cold air drainage. There is commonly in mountain valleys 
a down-valley wind which sets in shortly after sundown on calm 
days. This wind varies in velocity from a drifting of air to a 
strong wind. It forces aloft the warm air of the valley, which 
hangs for a time in mid-height in the valley. These nocturnal 
inversions, due primarily to the drainage of cool air from peaks, 
are common in the western arid part of the United States. For 
them one should sec the work of Shreve and other ecologists. 

The inversion due to adiabatic heating of a descending air 
current is clearly described by Humphreys. This type could be 
best demonstrated by downward currents on a slope which is 
steep in the upper reaches but which flattens out on approach to 
the valley floor. Initially, when the temperature of the free air 
everywhere over the valley decreases with elevation, the speed of 
the surface air down, at least the steepest portions of the cooling 
walls, is quite certain to be sufficient to make its dynamic gain in 
heat exceed its conduction loss, and therefore to cause its tem- 
perature to increase with descent. As the bottom of the valley is 
approached, however, the rate of vertical descent and the con- 
sequent dynamic heating become less and less, and so finally 



22 MOUNTAIN GEOGRAPHY 

cease altogether except in so far as there is drainage along the 
valley. At and near the bottom, then, where the dynamic heating 
is absent or small, the temperature of the surface and the adja- 
cent air necessarily decrease more or less rapidly. In a short 
while, therefore, the valley basin begins to fill with a river of 
cold air. 

If outside the mountain region there exists, perhaps for some 
days, a high pressure area of baric maximum, air currents are 
said to descend the barometric surfaces and then to experience an 
adiabatic heating as they meet and are compressed against the 
pool of cold, stagnant air of well enclosed valleys. Fruh has re- 
cently given us not only the data of such inversions but vivid 
descriptions of them. 

According to the meteorological description of such an inver- 
sion by Robert Billwiller, the heart of a Central-European maxi- 
mum hovered with 780 mm. over the Bohemian Forest. From 
December 16th to 28th the following daily averages resulted: 

Example of a Temperature Inversion 

Station Altitude Temperature 

Geneva 408 — 7.2 

Vitznau 445 — 4.9 

Glarus 471 —1 1.8 

Altsatten 478 — 12.0 

Neuchatel 488 — 10.6 

Trogen 892 — 5.4 

Elm 980 — 4.7 

Chaumont 1 128 1 .3 

Beatenberg 1 150 5.0 

Gabris 1250 2.7 

Rigi 179° °-7 

Great St. Bernhard 2478 — 4.9 

(Great St. Bernhard is obviously above the thermal belt.) 

That irradiation in itself is not the cause of the apparition is shown by 
the fact that during the night it is also warmer above, that the afore- 
stated temperatures in the morning occur before sunrise, that, for in- 
stance, the village of Elm was without sun from September 30th until 
March 12th, and yet from December 16th to 28th, 1879, had a daily 
mean higher by 7.1 than that of deeper located Glarus. Brightness and 
heat above the fog are the result of the descending air, the dynamic 
warming up. 1 

A work on inversion which will always stand as a classic is 

1 Fruh. 



MOUNTAIN TEMPERATURES 23 

H. J. Cox's study of North Carolina temperature. The thermal 
belts or verdant zones of North Carolina are the most striking 
with which the writer has personal experience. They occur in the 
Eastern Appalachian Mountains, in basins at the margin of the 
Asheville Plateau and, strangely enough, on the open exposed 
eastern flank of the mountains. Here the inversions are all noc- 
turnal inversions, accompanied frequently by an early morning 
fog. The vegetal response to this local prevention of frost is most 
striking. The writer has seen the valley at Tryon, North Caro- 
lina, brown with the dormant vegetation of winter. The moun- 
tain tops were white with snow, for here the mountains reach to 
some 1 350 meters above the sea. Along the side of the mountains 
between the brown and the white was a strip of green, the ver- 
dant zone, the result of inversion. Grapes of a hardy variety are 
grown in the valley, but at a great hazard from frost. In the ver- 
dant zone there is a more successful vignoble, 1 and here the sensi- 
tive Isabella grape has been grown for thirty years without danger 
from frost. 

Growing Seasons and Altitude 

The decreasing length of the growing season with increasing 
altitude is critical to plant zones. The following is a generalized 
table of the relation of growing season and altitude in Colorado: 2 

Growing Season and Altitude 

Zone elevation Length in days 

Below 5000 feet (1525 m.) 146 days 

5000-6000 " (1 525-1 830 m.) 138 " 

6000-7000 " (1830-2 1 35 m.) 113 " 

7000-8000 " (2135-2440 m.) 90 u 

8000-9000 " (2440-2745 m.) 56 " 

So generalized a table is not accurate for detail. A pronounced 
difference in length of growing season may occur between stations 
of approximately the same altitude. It is such exceptions to the 
generalization which devitalizes to a large extent the conception 
of true vegetation zones on mountains. The value of the con- 
ception of the zone is discussed in Chapter IV. Below are five 
Colorado stations whose variations in days between average 
frosts are disproportionally great as compared with their slight 

1 A French term for vine culture worthy of our adoption. 

2 Robbins. 



24 MOUNTAIN GEOGRAPHY 

differences in altitude. In the table above, it will be noted that 
there is on the average an increase of a day of growing season for 
every 44 feet in elevation. In the following table we note that the 
difference in elevation between Castle Rock and Cedaredge is 
55 feet, whereas the difference in length of the growing season is 

39 da Y s - 

Variations in Growing Season of Stations of 
Approximate Altitude 



Station 


Elevation 


Length in Days 


Castle Rock 


6200 feet (1897 m.) 


98 days 


Cedaredge 


. 6175 « (1883 m.) 


127 " 


Conejos 


. 7880 " (2303 m.) 


128 " 


Westcliffe 


• 7 86 4 " (2399 m -) 


IOO " 


Pagosa Springs 


7100 " (2166 m.) 


85 « 



Despite this table, there is some value in the conception of the 
general decrease of the growing season with altitude. The follow- 
ing table of well known Colorado farming communities illustrates 
a greater consistency of decrease of growing season. Lake City is 
the only exception to the comparative regularity of decrease. 

Regularity of Decrease of Growing Season 



Station 


Elevation 


Length in da 


Las Animas 


. . 3899 feet (1189 m.) 


161 days 


Fort Collins . . . 


• • 4985 " (1520 m.) 


144 " 


Glenwood Springs . 


5823 " (1776 m.) 


112 * 


Manos 


. . 6960 " (2123 m.) 


IO6 " 


Garnett 


• 7576 " (2310 m.) 


95 " 


Lake City 


. 8686 " (2649 m.) 


99 " 


Lake Moraine . . 


. . 10265 " (3138 m.) 


78 " 



This regularity is the more remarkable because these stations 
are not at increasing elevations upon a single slope but are widely 
scattered with a variety of exposures. 

We have also data upon growing season differences in our arid 
Southwest. The growing season at 8000 feet (2440 meters) in the 
Santa Catalina Mountains of Arizona is about half of that of the 
city of Tucson at 2360 feet (720 meters) at a distance of only 20 
miles (32 kilometers). 

Angot found, in the case of French observations, that if the 
dates of harvest were reduced to sea level the harvest was re- 
tarded a day for every 25 meters of elevation. Bene vent dis- 
covered the harvest isochrones in the French Alps to be oriented 



MOUNTAIN TEMPERATURES 



25 



west-southwest and east-northeast. In the Pyrenees the writer 
has seen at the same time rye ready for harvest and green rye 
higher up the slope. Elsewhere in the same range a 600 meters' 
difference in elevation meant 1 5 days' difference in harvest dates. 
This is at the rate of a day for every 40 meters. 

The variation between Angot's general estimate and the 
writer's particular estimate is not surprising when one considers 
the variations in climate other than altitude which should be con- 
sidered. Not the least important of the contributing factors is 
mass of mountain. Levainville, comparing the growing seasons 
of stations in the valley of Barcelonnette in the French Alps with 
stations of comparable altitude, but embedded in greater mass, 
demonstrated that greater mass of mountain lengthens growing 
season. 

Temperature Ranges 

It would appear from the low mean annual temperatures of 
mountain heights that we have there an essentially polar climate. 



35' 



30 



25 



































J 


v\ 
















f 
















^ 


J 






































■ 



<2 ia 24 6 12 18 24 6 12 16 

Fig. 5. Daily Ranges of Temperature and Altitude 

Daily temperature records at Zurich (493 m.) and Siintis (2500 m.) September 

2-5, 1895. (After Fruh.) 



Such is not the case. True, the Meije, a peak in Savoy of 3990 
meters, has a mean annual temperature of — 1 o degrees while the 
Siberian towns of Irkutsk and Verkoyansk have temperatures of 
— 1 1 and — 1 7 degrees respectively. The difference lies in the 



26 MOUNTAIN GEOGRAPHY 

seasonal temperature ranges. Verkoyansk, for example, has a 
highly continental mean annual range, whereas mountain peaks 
have ranges much more typical of marine climates, even to the 
retardation. We do not, then, in climbing a tropical mountain 
from jungle to eternal snow pass through climates equivalent to 
those from equator to pole. 

A still further difference from the polar climate is the intense 
insolation experienced in mountain climates. This intensity is a 
matter of great consequence to plant growth. The top of the 
Faulhorn in Switzerland, owing presumably to insolation, has on 
415 hectares 131 species of plants, while in the archipelago of 
Spitzbergen there are but 93. 

Both annual and diurnal ranges decrease with altitude. The 
following table illustrates the decreasing mean annual range: * 

Mean Annual Ranges in the Eastern Alps 

Height in Year in 

Place meters degrees Jan. July Range 

Bozen (Italy 290 11.7 0.0 22.5 22.5 

Brixen (Austria) 580 8.7 — 2.5 19.4 21.9 

Innsbruck (Austria) .... 600 7.9 — 3.3 17.8 21.1 

Sterzing 1000 6.2 — 4.4 16.3 20.7 

Schafberg 1780 1.6 — 5.1-Feb. 9.5 14.6 

St. Bernard 2470 —1.8 — 9.0 6.2 15.2 

Sonnblick 3 io 5 —6.6 — 13.3-Feb. 0.9 14.2 

Though the decrease of the diurnal range with altitude is not as 
great as that of the annual range, it is appreciable 2 (see also 
Figure 5): 

Decrease of Diurnal Range with Altitude 

Altitude in Amplitude of 

Place meters range in degrees Difference 

Burlington, Vermont 70 27.9 1 

Mount Washington, N. H 1916 22.9 / $' 



Catania, Sicily 30 16.2 

Mount Etna, Sicily 2 947 x o.8 



54 



Toulouse, France 190 16.4 1 

Pic du Midi, French Pyrenees 2877 14.1 J ** 

These, mountain stations are all relatively isolated peaks. The 
range at a given altitude is in part the function of the mass of 
mountain. Plateaus do not have the marked decrease of range 

1 Machatschek. 

2 Martonne. 



MOUNTAIN TEMPERATURES 27 

characteristics of isolated peaks. High valleys frequently have 
considerable range, particularly diurnal range, because of day- 
time heating through reflection and conduction of heat from en- 
closing valley walls and coldness at night due to cold air drainage. 
Indeed high valleys may have greater range than lowland sta- 
tions: * 

Temperatures in a High Valley 

Station Altitude Diurnal Range 

Geneva 407 m. 1 1 .6 degrees C. 

Chamonix (a high valley) io 34 x 4«2 

Mont Blanc 4810 3.5 

Values for gradients on concave and convex slopes are given by 
Fruh. Concave slopes are known as hatching ovens, fourneaux, by 
day, and cold holes by night. The concave slope concentrates 
heat waves, while the convex slope scatters them. Indeed, the 
topographic situation of a station is so great an influence in the 
range value that not every set of stations of increasing altitude 
shows decreasing ranges. 

Irregularity of Mean Annual Range Due to Situation 

Altitude in Mean annual 

Valley stations meters temperature C. Range 

Reichenhall 473 7.6 20.0 

Traunstein 597 6.7 20.1 

Pertenkirchen 715 6.5 18.9 

Tegernsee 742 7.1 17.4 

Obsertdork 820 5.8 19.2 

Mittenwald 910 6.0 17.1 

Peaks 

Wendelstein x 7 2 7 r *9 14..8 

Zugspitze 2964 ~~5- 2 l 3- 2 

Soil Temperatures 

It follows that if the upper levels of the atmosphere absorb 
little of the sun's radiant heat, rock exposures and soils at high 
levels consequently receive a proportionally large amount of 
heat energy. The rarity of the air likewise permits rapid radia- 
tion of heat from rocks. Simply stated, rocks and soil at high 
levels become excessively hot in daytime and cold at night. As 
air temperatures are low, though rock in the sun may be hot, 
rock in the shade is of much lower temperature than differences 

1 Vallot. 



28 MOUNTAIN GEOGRAPHY 

of soil temperature in sun and shade at low levels would in- 
dicate. 

This rapid heating and cooling accounts for the major part of 
weathering on mountain tops. Cracking and exfoliation are re- 
sponsible for the angular blocks of summit rocks, as for example 
the rocks of the much-photographed summit of Pikes Peak, 
Colorado. The effect of this rapid heating was well appreciated 
when the writer stood some distance from the sheer rock cliff 
known as the Royal Arches of Yosemite Park, California, as the 
morning's sun first struck the cliff. Pieces of rock were flaked 
from the surface and whirled through the air until they sounded 
like bullets. 

The soil temperatures of a sunny slope may be higher than soil 
temperatures of a valley station, for not uncommonly the high 
slope makes a greater angle of incidence with the sun's rays than 
does the valley floor. Though soil heat is derived largely from 
insolation rather than from air heat, yet, because radiation is so 
active in the rarer air levels, the average temperature of the soil 
decreases with altitude. As the vertical soil temperature gradient 
is less than that of the air the difference between the two is pro- 
gressively greater as one ascends: * 

Difference in Soil and Air Temperatures with Altitude 

Difference of soil and 
Altitude in meters air temperatures in degrees C. 

6oO O.5 

90O I .O 

I200 1.3 

I5OO 1.7 

180O 2.0 

2IOO 2.3 

24OO 2.5 

270O 2.7 

3OOO 2.9 

The mean difference between the soil and air temperatures as 
observed in the Tirol is 1.5 degrees at 1000 meters, 1.7 degrees at 
1300 meters, and 2.4 degrees at 1600 meters. The difference 
varies, however, with exposure, water content of the soil, slope, 
and latitude. In the National Park of Switzerland at 2200 meters 
the soil temperature has been observed to be 3.6 times that of the 
air. The following table illustrates the differences between the 

1 Jenny. 



MOUNTAIN TEMPERATURES 29 

mean and maximum soil temperatures at Bagneres-de-Bigorre 
and the Pic du Midi in the French Pyrenees: 

Maximum and Minimum Soil Temperatures 

Bagneres Pic du Midi 

511 meters 2877 meters Difference 

Mean air temperature 22.3 10. 1 —12.2 

Mean soil temperature 36.1 33.8 — 2.3 

Maximum air temperature 27.1 13.2 —I 3-9 

Maximum soil temperature 50.3 52.3 2.0 

The temperature of the soil, dependent as it is on insolation, 
varies with exposure. The following observations on soil were 
made at the depth of 80 centimeters in Austria: 

Soil Temperatures and Exposures 

Exposure Inn valley (600 meters) Gschnitz valley (1340 m.) 

N 9.5 5.1 

NE 10.6 5.5 

E 1 1.3 5-9 

SE 12.6 7.5 

S 12.6 7.8 

SW 12.7 7.4 

W . . 12.2 7.4 

NW 10.2 9.5 

Mean 1 1.5 6,7 

Range 3.2 2.7 

The migration of the maximum soil temperature from southwest 
in winter to southeast in summer is probably due to the diurnal 
variation in cloudiness. 

The significance of soil temperatures at high levels is even more 
outstanding than at low levels. This is brought out by Shreve. 
It makes possible the promotion of root growth at levels where 
air temperatures would be prohibitive to the germination of 
seeds. Radiation strongly affects air temperatures near the 
ground, which accounts, in part, for the low branching of bush 
and tree growth. Moreover, the local warming of the air causes 
minor convection currents. The rate of evaporation on the 
sunny slope is less favorable for plants than that of the shady 
slope. Soil temperatures influence the length of the growing 
season and the duration of snow cover. All this Shreve supports 
with instrumental observations. One should also consult the 
work by Jenny. 



30 MOUNTAIN GEOGRAPHY 

BIBLIOGRAPHICAL NOTES 

General References on Mountain Climate 

The great book on this subject is Hann, revised by Knoch. It is a 
complete treatise in subject matter, taken, unfortunately, almost en- 
tirely from German source material. The Bibliographical Note on 
regional works at the end of Chapter VII should also be consulted. 

Bach, Hugo. Das Klima von Davos. Zurich, 1907. A high climate fully 

treated. 
Benevent, Ernest. Le climat des Alpes francaises; Memorial de F Office 

National Meteor ologique de France. Paris, 1926. A regional treatise of 

general application. 
Blanchard, Raoul. "Le contraste climatique entre Vercors et Diois," 

in Recueil des travaux de PInstitut de geographie alpine, vi (191 8), pp. 

427-446. Involves fundamental principles. 
Davis, W. M. "A Speculation in Topographical Climatology," in 

American Meteorological Journal, xii (1896), pp. 372-381. An early 

American article. 
Davis, W. M. "Mountain Meteorology," in Appalachia, iv (1885), pp. 

225-244, 327-35 - 
Dorno, Carl. " Klimatologie des Hochgebirges," in Verhandlungen der 

Klimatologischen Tagung in Davos, 1925, pp. 130-140. Medical 

climatology. 
Exner, Franz. "Klima der Alpen," in Die bsterreichischen alpen, ed. by 

Hans Leitmeier. Leipzig and Vienna, 1928, pp. 165-175. General. 
Ficker, Heinz von. Klimatographie von Tirol und Vorarlberg. Vienna, 

I9 ° 9 * 
Hann, Julius. Handbuch der Klimatologie, 4. Aufl. by Karl Knoch. Vol. 

i. Stuttgart, 1932. The great book on the subject. 
Hann, Julius. Handbook of Climatology, part i, trans, by R. DeC. Ward. 

New York, 1903. 
Huber, Anton. Das Klima des bayerischen Alpenanteiles und seines Vorlandes. 

Munich, 1929. (Veroffentlichungen der Gesellschaft fur bayerische Landes- 

kunde, 5.) 
Humphreys, W. J. Physics of the Air, 2d ed. New York, 1929. A book 

of prime importance for the theorist. 
Kendrew, W. G. Climate. Oxford, 1930. 
Knoch, Karl. Klimakunde von Sudamerika. Berlin, 1930. {Handbuch der 

Klimatologie, ed. by Wladimir Koppen and Rudolf Geiger, vol. ii.) 

Contains all the text and references needed for South American 

studies. 
Martonne, Emmanuel de. Traite de geographie physique, 4 e ed. Vol. i 

(Paris, 1925), pp. 307-33 1 - 



MOUNTAIN TEMPERATURES 31 

Maurer, Julius, Billwiller, Robert, and Hess, Clemens. Das Klima der 
Schweiz- Frauenfeld, 1909-10. 2 vols. 

Miller, A. A. Climatology. London, 193 1. Immethodical but readable 
material. 

Peattie, Roderick. "Les apports de PAmerique a la climatologie de 
montagne," in Melanges geographiques offer ts a Raoul Blanchard (Gre- 
noble, 1932), pp. 467-479. Also in Revue de geographie alpine, xx 
(i93 2 ) '> PP- 253-266. 

Rotch, A. L. "Mountain Meteorology," in American Meteorological 
Journal, viii (1891), pp. 145-158, 193-21 1. Of historical importance 
only. 

Schlagintweit, Hermann and Adolph von. Untersuchungen iiber die 
physicalische Geographie der Alpen. Leipzig, 1850. An early treatise. 

Shaw, Sir Napier. Manual of Meteorology. Vol. iii, Cambridge, Eng- 
land, 1930. 

Shreve, Forrest. The Vegetation of a Desert Mountain Range as Conditioned 
by Climatic Factors. Washington, 191 5. (Carnegie Institution of 
Washington, Publications, 217.) Excellent for mountains in an arid 
region. 

Stone, R. G. "The History of Mountain Meteorology, in the United 
States and the Mount Washington Observatory," in Transactions of 
the American Geophysical Union, 15th Annual Meeting, 1934, pp. 124— 
133. With an excellent bibliography. Or see "Die Entwicklung 
der amerikanischen Bergobservatorien und das derzeitige Netz von 
Bcrgstationen in der Vereinigten Staatcn von Amerika," in Jahres- 
bericht of the Sonnblick-Verein, xliii (1934), pp. 11-30. 

Supan, Alexander. Grundzuge der physischen Erdkunde, 7. Aufl. by Erich 
Obst. Berlin, 1927. 2 vols. 

Ward, R. DeC. Climate, Considered especially in Relation to Man. New 
York, 1908. 2d ed., revised, 191 8. 

Ward, R. DeC. The Climates of the United States. Boston, 1925. 

Ward, R. DeC. "A Visit to the Highest Meteorological Station in the 
World," in Monthly Weather Review, xxvi (1898), pp. 150-152. 

Wegener, Alfred. Thermodynamik der Atmosphare. Leipzig, 191 1. 

Insolation 

One should consult the general references of the previous biblio- 
graphical note. 

Angstrom, Anders. " On the Atmospheric Transmission of Sun Radia- 
tion," in Geografiska Annaler, xi (1929), pp. 156-166; xii (1930), pp. 
130-159. Has a good bibliography. 

Church, J. E. "Summit Temperatures in Winter in the Sierra Ne- 
vada," in Appalachia, xi (1905-08), pp. 239-248. 



32 MOUNTAIN GEOGRAPHY 

Garnett, Alice, " Insolation, Topography, and Settlement in the Alps," 
in Geographical Review, xxv (1935), pp. 601-617. 

Geiger, Rudolf. Das Klima der bodennahen Lujtschicht. Braunschweig, 
1927. 

Harshberger, J. W. "Alpine Fell-Fields of Eastern North America," in 
Geographical Review, vii (1919), pp. 233-255. 

Harshberger, J. W. " Slope Exposure and the Distribution of Plants in 
Eastern Pennsylvania," in Bulletin of the Geographical Society of Phila- 
delphia, xvii (191 9), pp. 53-6i. 

Kendrew, W. G. Climate. Oxford, 1930. Chap, xxxviii. 

Kimball, H. H. " Observations on the Increase of Insolation with Ele- 
vation," in Bulletin of the Mount Weather Observatory, vi (191 3-14), 
pp. 107-1 10. 

Kolhorster, W., and Salis, G. von. "Variation of Penetrating Radia- 
tion on the Jungfrau," in Nature (London), cxviii (1926), p. 518. 

Okada, T., and Yosida, Y. " Pyrheliometric Observations at the Sum- 
mit and at the Base of Mount Fuji," in Bulletin of the Central Mete- 
orological Observatory of Japan, no. 3. Tokyo, 1910. 

Bibliographies in Monthly Weather Review, lv (1927), pp. 163-166, 168- 
169. 

Temperature and Altitude 

Ahlmann, H. W., and Eythorsson, J. "Introductory Survey of the 
Temperature Conditions in the Horung Massif during the Sum- 
mers of 1923-1926," in Geografiska Annaler, ix (1927), pp. 13-21. 

Anfossi, G. "Volumetric de la Corse," in Recueil des travaux de Pin- 
stitut de geographie alpine, vi (191 8), pp. 27-69. The measurement of 
mass of mountain. 

Anfossi, G. "Volumetria della Sardagna," in Memorie Geografiche, sup- 
plement alia Rivista Geografica Italiana, ix (19 15), pp. 181-235. The 
measurement of mass of mountain. 

Angot, Alfred. " La temperature de la France," in Annales de geographie, 
xiv (1905), PP. 296-309. 

Angot, Alfred. "Sur la decroissance de la temperature dans Pair avec 
la hauteur," in Comptes rendus hebdomaires des seances de V Academic de 
Sciences. Paris, cxv (1892), pp. 12 70-1 2 73. 

Backman, Allan. "L'influence de Paltitude et de la position geogra- 
phique sur la temperature de Pair dans la region sud-ouest de la 
Suede," in Geografiska Annaler, ix (1927), pp. 133-141. 

Budel, Anton. "Die Zugspitzbahn," in Deutsches Meteor ologischen 
Jahrbuch, 1929, pp. E 1-30, and 1930, pp. E 1-5. 

Hann, Julius. "Die Mittlere Warmevertheilung in den Ostalpen," in 
Zeitschrift des Deutschen und Osterreichischen Alpenvereins, xvii (1886), 
pp. 22-94. A classic. 

Hann, Julius. "Die Temperaturverhaltnisse der osterreichischen 



MOUNTAIN TEMPERATURES 33 

Alpenlander," in Sitzungsberichte of the Vienna Academy, math.- 

nat. Klasse, xc, 2 (1884), pp. 585-683; xci, 2 (1885), pp. 4°3~453; 

xcii, 2 (1885), pp. 33-198. 
Hann, Julius. "Zur Meteorologie des Sonnblickgipfels," in J^eitschrift 

des Deutschen und Osterreichischen Alpenvereins, xx (1889), pp. 71-93. 
Henry, A. J. "Daily Changes in Temperature up to 4,000 Meters," in 

Bulletin of the Mount Weather Observatory, v (1912— 13), pp. 1— 18. 
Henry, A. J. " Variations of Temperature and Pressure at Summit and 

Base Stations in the Rocky Mountain Region," in Bulletin of the 

Mount Weather Observatory, iii (1910-11), pp. 201-225. 
Henry, A.J. " Variations of Temperature at Summit and Base Stations 

in the Central Rocky Mountain Region," in Bulletin of the Mount 

Weather Observatory, iv (191 1— 12), pp. 103— 114. 
Henry, A. J. "The Temperature at Mount Weather and Adjacent 

Valley Stations," in Bulletin of the Mount Weather Observatory, iv 

(1911-12), pp. 310-341. 
Henry, A. J. "Vertical Temperature Gradients between Mount 

Weather, Va., and Valley Stations," in Bulletin of the Mount Weather 

Observatory, vi (1913-14), pp. 35-37. 
Hettner, Alfred. Die Kordillere von Bogota. Gotha, 1892. {Petermanns 

Mittcilungen, Erganzungsheft Nr. 104.) 
Huber, Anton. "Das Klima dcr Zugspitze," in Beobachtungen der Me- 

teorologischen Stationen im Konigreich Bayern, xxxv (1913), pp. L 3— 

L62/ 
Imhof, Eduard. "Die Waldgrenze in der Schweiz," in Beitrage zur 

Geophysik, iv (1900), pp. 241-330. Forests and mass of mountain. 
Jegcrlehner, J. "Die Schneegrenze in den Gletschergebieten der 

Schweiz," in Beitrage zur Geophysik, v (1903), pp. 486-568. Snow 

and mass of mountain. 
Liez, H. "Die Verteilung der mittleren Hohe in der Schweiz," in 

Jahresbericht der Geographischen Gesellschaft von Bern, xviii (1900-02), 

pp. 1-38. 
Marek, Richard. "Beitrage zur Klimatographie der oberen Wald- 
grenze in den Ostalpen," in Petermanns Mitteilungen, lvi, 1 (1910), 

pp. 63-69. 
Qucrvain, Alfred de. "Die Hcbung der atmospharischen Isothermen 

in den Schweizer Alpen und ihre Beziehung zu den Hohengrenzen," 

in Beitrage zur Geophysik, vi (1904), pp. 481-533. 
Schalgintwcit, Hermann and Adolph von. Neue Untersuchungen iiber die 

physicalische Geographie und die Geologie der Alpen, Leipzig, 1854. 
Shreve, Forrest. " Conditions indirectly Affecting Vertical Distribution 

on Desert Mountains," in Ecology, iii (1922), pp. 269-274. 
Somervell, T. H., and Whipple, F.J. W. "The Meteorological Results 

of the Mount Everest Expedition," in Qiiarterly Journal of the Royal 

Meteorological Society, Iii (1926), pp. 131-143. 



34 MOUNTAIN GEOGRAPHY 

Vallot, J. " Variation de la temp6rature, de la pression, et de la vapeur 
d'eau au Mont Blanc et aux stations inferieures, d'apres les observa- 
tions de 1887," in Annales de VObservatoire du Mont Blanc, i (1893), pp. 
25-45- 

Inversions of Temperature 

For theory one should turn to the handbooks of Hann, Shaw, and 
Humphreys. For description, see the work on Switzerland by Friih. 
Abbe, Cleveland. "Thermal Belts, Frostless Belts, or Verdant Zones," 

in Monthly Weather Review, xxi (1893), p. 365. 
Andre, Charles. Influence de V altitude sur la temperature. Lyons, 1888. 
Blache, Jules. Les massifs de la Grande Chartreuse et du Vercors. Grenoble, 

1 93 1. 2 vols. Vol. i, pp. 415-419. 
Blache, Jules. " Notes sur les conditions de l'inversion de temperature 

dans la region du Villard-de-Lans," in Revue de geographie alpine, xx 

(1932), pp. 36i-37 - 
Brown, W. P. "Winter Temperatures on Mountain Heights," in 
Quarterly Journal of the Royal Meteorological Society, xxxvi (1910), pp. 

I7-I9- 
Chickering, J. W. " Thermal Belts," in American Meteorological Journal, 

i (1884-85), pp. 213-218. 

Church, J. E. ''Summit Temperatures in Winter in the Sierra Ne- 
vada," in Appalachia, xi (1905-08), pp. 239-248. 

Clarke, W. T. " Peaches and Climate," Monthly Weather Review, xxxviii 
(1910), p. 1740. 

Cox, H. J. "Weather Conditions and Thermal Belts in the North 
Carolina Mountain Region and their Relation to Fruit Growing," 
in Annals of the Association of American Geographers, x (1920), pp. 57-68. 

Cox, H. J. "Thermal Belts and Fruit Growing in North Carolina," in 
Monthly Weather Review, supplement no. 19 (1922), pp. 1-98. 

Dauzere, Camille. "Sur les inversions de la temp6rature," in Comptes 
rendus hebdomadaires des seances de V Academic des Sciences, Paris, clxxxii 
(1926), pp. 978-980. 

Davis, W. M. "Types of New England Weather," in Annals of the 
Astronomical Observatory of Harvard College, xxi, 2 (1889), pp. 1 16-137. 

Friih, Jacob. Geographie der Schweiz- St. Gallen, 1929-33. 2 vols. 
Vol. i. 

Henry, A. J. "Temperature Inversions at the Mount Weather Ob- 
servatory," in Bulletin of the Mount Weather Observatory, i (1908), pp. 
143-160. 

Hutt, W. N. "Thermal Belts from the Horticultural View-Point," in 
Monthly Weather Review, supplement no. 19 (1922), pp. 99-106. 

Kerner, A. "Die Entstehung relativ hoher Lufttemperaturen in der 
Mittelhohe der Thalbecken der Alpen im Spatherbste und Winter," 



MOUNTAIN TEMPERATURES 35 

in geitschrift der Osterreichischen Gesellschqft fur Meteor ologie, xi (1876), 
pp. 1-13. 
MacDougal, D. T. " Influence of Inversions of Temperature, Ascend- 
ing and Descending Currents of Air, upon Distribution," in Bio- 
logical Lectures from the Marine Biological Laboratory of Woods Hole, 
i8gg (Boston, 1900). 

Soil Temperatures 

Buhler, A. " Einfluss der Exposition und des Neigungswinkels auf die 
Temperatur des Bodens," in Mitteilungen der Schweizerischen Central- 
anstaltfiir das forstliche Versuchswesen, iv (1895). 

Jenny, Hans. " Hochgebirgsboden," in Handbuch der Bodenlehre, ed. by 
Edwin Blanck, iii (Berlin, 1930), pp. 96-118. Excellent bibliog- 
raphy. 

Kerner, A. " Uber Wanderungen des Maximums der Bodentempera- 
tur," in Zeitschrift der Osterreichischen Gesellschqft fur Meteor ologie, vi 
(1871), pp. 65-71. 

Kerner von Marilaun, Fritz. " Die Anderung der Bodentemperatur mit 
der Exposition," in Sitzungsberichte of the Vienna Academy, math.- 
nat. Klasse, c, 2a (1891), pp. 704-729. 

Shreve, Forrest. "Soil Temperature as Influenced by Altitude and 
Slope Exposure," in Ecology, v (1924), pp. 128-136. 



CHAPTER II 
HUMIDITY AND PRECIPITATION 

Humidity and Evaporation 

THE decrease of absolute humidity with altitude is more rapid 
than the diminishing of other atmospheric elements here con- 
sidered. The following table illustrates the theoretical relative 
decrease of vapor tension and atmospheric pressure with increas- 
ing altitude: x 

Decrease of Vapor Tension and Air Pressure with Altitude 

Altitude in meters Vapor tension Air pressure 

O I .OO 1 .00 

IOOO O.73 O.88 

QOOO O.49 O.78 

3OOO O.35 O.69 

4OOO O.24 O.61 

5OOO O.17 O.54 

60OO O.I2 O.47 

This table is merely a slide rule by which vapor tensions may be 
generalized. Thus, if the vapor tension at 10 millimeters is found 
at sea level, the tension at 4000 meters would be 2.4 millimeters. 

The actual absolute humidity found at elevations in mountains 
varies to such a degree as to defy generalization. It is, of course, 
a matter of weather and such permanent influences as exposure, 
angle of slope, character of soil, condition of ground water, and 
mass of mountain. It has its diurnal and seasonal altitudinal dis- 
tribution. In the mornings the humidity of the air is great in the 
valleys as compared with the upper reaches. By afternoon con- 
vection currents have distributed this moisture in part to upper 
levels. The low absolute humidity of a high level station aug- 
ments the intensity of insolation there and reduces the heat 
absorbing quality of the air. 

Relative humidity and increasing altitude have no consistent 
relationship. Relative humidity is the ratio between the abso- 

1 Hann. 



HUMIDITY AND PRECIPITATION 37 

lute humidity of the moment and the saturation point at the 
existing temperature, and is therefore dependent upon the abso- 
lute humidity and the temperature. The irregularity of the verti- 
cal temperature gradient is more or less to be counted upon. 
The irregularity of the absolute humidity gradient has just been 
referred to. Relative humidity is the result, then, of two varia- 
bles and cannot be generalized or reduced to a formula. 

Evaporation and its significance are particularly difficult to 
treat because so little is known of the subject in high altitudes in 
its quantitative sense. In most high areas the potential evapora- 
tion is actually greater than precipitation. The effectiveness of 
humidity depends upon the potential rate of evaporation, which, 
in turn, depends upon the barometric pressure, the amount of 
vapor already in the air, the temperature of the soil, the tem- 
perature of the air, the velocity of the wind, and the character of 
the soil and vegetal covering. The less the barometric pressure 
and the less the water vapor content of the air, the more rapid the 
evaporation. All air constituents, and especially water vapor 
tension, constitute a hindrance to the release of gas molecules 
from water, the process which we know as evaporation. 

The temperature of the air at high elevation is low, except in 
the shallow zone in immediate contact with the rock. On the 
other hand, the temperature of the soil is primarily a question of 
its physical character, its water content, its exposure, angle of 
slope, and vegetal covering. The physical character of the soil 
obviously influences soil temperatures. The scarcity of Alpine 
vegetation, though it decreases transpiration, increases evapora- 
tion of ground water because of the greater exposure of barren 
ground to the sun. The importance of soil temperatures is illus- 
trated photographically and with text by R. R. Piatt in John- 
son's Peru from the Air (1930). On page 33 the legend of an illus- 
tration runs: " There is abundant rainfall here, but so intense is 
the heat of the afternoon sun that the northern and western 
slopes of the hills that border the valley do not retain sufficient 
moisture to support tree growth. Only in the gullies and on the 
spurs that are shaded from the afternoon sun is there forest." In 
the writer's New College Geography, page 465, appears a picture 
from the collection of E. M. Spieker. This shows a forested slope 
facing a barren slope. The legend reads: "This scene is taken on 
the Wasatch Plateau at an elevation of 8000 feet (2440 meters) in 
the valleys to 9000 feet (2745 meters) at the summits. The bar- 



38 MOUNTAIN GEOGRAPHY 

ren slope faces south. Here the sun's rays are hot enough to 
cause evaporation to an extent which prevents tree growth." 

The velocity of the wind is critical in evaporation. The in- 
creased speed of wind with altitude is shown by the following 
table of averages for three stations: 

Relative Average Wind Velocities 

Altitude Relative speed 

Kremsmunstcr 390 m. 3.5 

Saentis 2500 m. 7.5 

Sonnblick 3100 m. 9.3 

The effect upon the Alpine plants is to eliminate all but those of a 
xerophytic nature. Is not the Edelweiss, famed flower of the 
heights, of this type? Strength of evaporation can be measured 
by the climber through the rapidity of the drying of perspiration 
and the great thirst which are experienced on mounting. Among 
high peaks dead animals are found mummified. Air-dried meat 
is a provincial dish in the high Engadine. Great and rapid 
evaporation is one of the most characteristic climatic conditions 
of high elevations. 

Precipitation 

A primary concern in any discussion of precipitation at high 
altitudes is the paucity and inaccuracy of observations. In the 
nature of things, a rain gauge is not as automatic or accurate as a 
thermograph. In any case, year round records of high level 
precipitations are few. The writer is convinced that recorded 
totals of mountain rainfalls are frequently short of the true total. 
In spite of counter devices, wind-driven rain is not recorded as 
completely by the gauge as where there is relative calm. A high 
wind during a rain storm in mountains is common. But still 
greater inaccuracy exists in the failure of the rain gauge to record 
precipitation such as occurs on rocks and plants from fog or 
cloud driven against the cold mountain side. A support of this 
doubt lies in the fact that relatively small drainage basins in high 
mountains produce streams disproportionately large as com- 
pared with streams of similar drainage area on lowlands. 

Generally speaking, mountains have higher precipitation 
values than surrounding plains. This is due to the uplift and 
cooling of winds. Even winds of slight relative humidity will 
reach a saturation point if sufficiently cooled. This uplift of the 



HUMIDITY AND PRECIPITATION 



39 



wind may involve a considerable zone of the air. The depth may 
be so great that the flexing of the zone may cause uplift of air 
even over a piedmont area of no great relative elevation. Moun- 
tain piedmonts are more rainy than adjacent plains. Thus the 
depth of the monsoon wind is considerable, with a resulting 
heavy rainfall on the piedmont of the Himalayas disproportionate 
to the slightly greater elevation of the piedmont over the plains. 1 

Influence of Proximity of Mountains on Precipitation of 
Piedmont Stations 

Station Dacca Bogra Mymensing Silhct 

Distance from Himalayas in kilometers 161 96 48 32 

Rainfall in centimeters 191 231 274 380 

The increase of precipitation with altitude is characteristic 
only of the windward slope. The converse is true of the lee 
slope. On the lee side the wind is descending and therefore 
warming. As the descending wind warms, rather than permitting 
condensation of moisture, it is constantly acquiring a greater and 
greater capacity for absorbing moisture. There is on the lee 



SIERRA NEVADA 

51.18 In. 
47.02 in. ^47.74 «> 



GREAT BASIN 




SACRAMENTO VALLEY 

19.78 In 019.6O in 3 



4077 



Fig. 6. Topography and Precipitation, American Sierra Nevada 

Note that the summit station is above the zone of maximum precipitation. Reno 
lies in the * rain shadow.' 

slope a region drier than the more general weather controls of 
that part of the world would justify. This region is said to be in 
the 'rain shadow 5 of the mountains. It should be noted that the 
lee slope of the summit may have considerable rain as a carry- 
over from the rain on the windward slope. 

The factors involved in the increase of precipitation with alti- 
tude are the amount of moisture in the air in terms of absolute 
and relative humidity, the angle of the mountain slope, the 

1 de Martonne. 



40 MOUNTAIN GEOGRAPHY 

height of the range, and the vertical temperature gradient. What 
actual precipitation occurs on a mountain range depends upon 
climatic location, orientation of the mountain crest, relative 
height of the range above the plain, and the angle of the slope 
of the mountain. The rate of precipitation from sea level to a 
zone of maximum precipitation has been computed by formulae 
by a number of scientists. McAdie has worked out an elaborate 
formula for his California study. De Martonne quotes Huber, 
who has devised a formula that includes altitude and angle of 
slope. Precipitation increases with altitude until the zone of 
maximum precipitation is reached. The following table, from 
California, illustrates the increase rate, the zone of maximum 
rainfall, and the rain shadow: 

Precipitation in California and the Sierra Nevada 
(See also Figure 6) 

Station Elevation Relative amounts 

Oakland (on San Francisco Bay) 36 feet 1 00 

Sacramento (in California Valley) 7 1 " 80 

Rocklin 249 " 121 

New Castle 956 a 140 

Auburn 1 363 " 165 

Colfax 242 1 " 237 (max.) 

Gold Run (maximum zone) 3222 " 226 

Summit (highest elevation) 7037 " 223 

Boca (rain shadow) 553 1 " 83 

Reno, Nevada (rain shadow) 4484 " 35 

The Sierra Nevada is an example of a range with crest line 
normal to the prevailing wind direction. All mountains which 
are barriers to wind progress show similar precipitation condi- 
tion. Mountain ranges whose crest lines are parallel to the wind 
direction show little or no increase of precipitation with altitude. 
Below is a table to illustrate the relationship of altitude and pre- 
cipitation in the French Alps. Examples from three slopes are 
given on the opposite page. 1 

Hann gives formulae for the relations of rainfall and altitude. 
Blache analyzes the inaccuracy of records. Shreve discusses the 
irregularity of decrease on desert ranges of Arizona. The irregu- 
larities are those of observation, always found on mountains, and 
the irregularities due to slope and exposure. 

1 Blache. 



HUMIDITY AND PRECIPITATION 41 

Rainfall and Altitude in the Chartreuse and Vercors 

Station elevation Precipitation 

215 m. I 1 53 mm. 

275 1282 

1025 1328 

I200 I446 

4IO I707 

643 I947 

27O I I 46 

I OOO I 500 

The zone of maximum precipitation is of prime significance in 
consideration of mountains and their human relations. There are 
some authorities who argue that forest limits are controlled to a 
greater or less degree by this upper limit of the zone of maximum 
precipitation. Certainly decrease of moisture is one factor in the 
upper limit of tree distribution. How important it is depends in 
part upon the comparative aridity of the region. Not enough 
measurements have been made to determine properly the zone of 
maximum precipitation. Hann has stated that such a zone in the 
Alps approximates 2000 meters in elevation. Benevent, dividing 
the northern section of the French Alps into the Fore-Alps, Cen- 
tral Alps, and Intra-Alpine Zone, places the maximum precipi- 
tation at 2000 meters, 2500 meters, and 2500 to 3000 meters 
respectively. Joseph Vallot, who established the observatory on 
Mont Blanc, makes the assertion that the summit of Mont Blanc 
has less precipitation than Montpellier, a town on the dry Medi- 
terranean coast of France. The maximum height of the zone in 
the Pyrenees is not definitely ascertained, and one authority be- 
lieves it to be higher than the usually accepted figure of 2000 
meters. The writer's observation in Andorra, in the heart of the 
range, is that the summit plateau at 2000 to 2600 meters is rela- 
tively dry as compared with the slopes. The relative aridity is a 
definite barrier to beech trees and rye at these higher levels. 
There is danger, however, in judging rainfall of higher altitudes 
by the character of plant cover, since high winds and strong in- 
solation there counteract the effectiveness of precipitation to a 
high degree. It must be noted, nevertheless, that the shepherds 
in the Tian Shan and the Pamirs actually drive their flocks to 
higher levels in winter than in summer in order to avoid the zone 
of maximum winter snowfall. 



42 MOUNTAIN GEOGRAPHY 

Mass of mountain once again deserves consideration. It is a 
matter of common observation that the greater the mass the 
higher the level of the clouds. The cloud zone on Mont Blanc is 
said to be 3000 meters, and in the Fore-Alps 2000 meters. The 
greater the cloudiness of a zone the less the evaporation and the 
more effective the rainfall. 

One looks to the lowering of cloud zones in winter as the rea- 
son for slight winter precipitation in high altitudes. At Davos, 
at 1560 meters, January is the driest month (45.7 mm.) and July 
and August the wettest months (127.7 nim.). This is a range of 
82 mm. amplitude. It must be remembered that this is true in 
spite of the fact that the rainy season of Switzerland is in winter. 
Bach, who has demonstrated the above fact, finds the same 
anomaly in the case of other high level stations. It must not be 
assumed, however, that the annual average decrease of rainfall 
with altitude would show the same curve. The low winter level 
of maximum precipitation is of short duration. 

The most common cause for the difference in rainfall totals of 
two near stations of like altitude is the matter of exposure. The 
following table is a rainfall cross-section of the Vosges Mountains 
of France. Nancy on the windward side and Colmar on the lee- 
ward slope have the same altitude: * 

Precipitation of Two Flanks of the Vosges 

West (windward) East (leeward) 

Nancy Mirccourt Epmal Le Syndicat Wcsserlmg Thaun Colmar 

Altitude (m.) 200 279 339 620 437 238 200 

Precipitation (mm.) ... 786 881 950 1374 1208 932 479 

One of the most extraordinary examples of exposure is on a 
mountain on an island in the north of the Hawaiian group. Here 
the rainfall on the windward side has been measured to more 
than 450 inches in a year, whereas seventeen miles to the lee- 
ward the rainfall is but twenty-two inches. 

The influence of exposure is well illustrated by the French 
Alps. The driest section of this group is the Briangonnais, which 
owes its significant aridity to protection. Of the French Alps, the 
Savoian section is in the belt of westerly winds. The Provencal 
section is in the Mediterranean zone. High crests mark the 
division between the two sets of valleys. Savoy is decidedly moist 
as compared with Provence. The following table emphasizes 

1 de Martonne. 



HUMIDITY AND PRECIPITATION 43 

this contrast, showing that the Provengal station, though at a 
lesser altitude than the Savoian station, has the greater rainfall. 

Exposure and Precipitation in the French Alps 

Monts du Vercors Alpcs-Maritimes 

(Savoy) Lcntc (Provence) Thorcne 

Altitude 1080 m. 1250 m. 

Total Precipitation ^39 mm. 1 171 mm. 

Exposure to wind is so important in rainfall totals that in a 
sense the word wind is often equivalent to weather. Le vent is the 
same as Regenluft, and Regentwcr is the weather wind. South of 
the Wetter horn (Bernese Oberland) is the Wetterlueche or the gap 
through which the rain wind comes. The location of a station in 
line with prevailing winds coming through a gap or pass ac- 
counts for many local variations of rainfall within a small area. 

Mountain barriers cause comparative aridity in their lee. Such 
leeward stations, relatively dry for their altitude, are known as 
dry holes. A well known example is the region of Sion in the 
Upper Rhone Valley. Tremendous barriers there keep out 
weather winds and rain. On the other hand, a comparatively 
slight topographic difference will account for paucity of rain. 
The towns of Prades and Villefranche, in the valley of the Tet, 
Eastern French Pyrenees, are dry holes, yet the barriers against 
rain weather are merely of foothill proportions. Exposure, caus- 
ing contrasts of precipitation, is one of the most decisive of climate 
controls in mountains. 

Dry Holes in the Eastern French Pyrenees 

Station Altitude in meteis Rainfall in millimeters 

Perpignan 31 554 

Vinca 259 612 

Prades 354 500 

Villefranche 889 465 

Olette 1 1 33 560 

Theres 1 1 37 597 

Mont Louis 1586 827 

Snow 

Snow on mountains must be considered from the point of view 
of quantity and frequency of fall and duration of mantle. The 
factors which control these considerations are total precipitation, 
altitude, slope, exposure, and evaporation. Precipitation varies 



44 MOUNTAIN GEOGRAPHY 

greatly with the exposure. The percentage of the precipitation 
which is snow is more or less directly in relation to the altitude 
and latitude. For two stations of the same latitude and same 
general climatic situation, the greater the altitude the greater the 
snowfall, until the zone of maximum snowfall is reached. 

Number of Days of Snowfall at Different Altitudes in Bavaria 
(The Roman numerals indicate months) 

1888-89 1890-91 

Station Alt. Begin. End Days Begin. End Days 

Miinchen 526 21 II 10 III 18 25 XI 22 II 90 

Rosenheim 449 3 II 24 III 50 25 XI 5 III 101 

Mieabach 717 3 II 12 IV 69 25 XI 10 III 106 

Onsersdorf 842 3 II 20 IV 77 24 XI 22 IV 150 

Peissenberg 994 3 II 21 IV 78 25 XI 17 III 113 

Wendelstein 1730 3 II 22 IV 79 17 X 8 V 204 

There are differences in factors influencing the date of the first 
day of snow cover and the last day. The first appearance of snow 
on all but the higher reaches is frequently the result of general 
weather conditions of the area. Therefore altitudes of a con- 
siderable range of elevation will experience snow on the same 
day. The duration of snow in the spring time is, however, the 
result more largely of weather factors. Most of these local cli- 
matic conditions have direct relation to altitude. Therefore the 
disappearance of snow begins at lower levels and gradually 
mounts higher. The exceptions are largely those of exposure. A 
shady slope, especially if forested, may hold snow far out of sea- 
son. A well insolated mountain slope may lose its snow before a 
valley bottom. Nevertheless the number of days of snow cover is, 
up to the zone of maximum snowfall, a rough measure of alti- 
tude. 

The variation of total days of snow cover in the two years pre- 
sented in the above table is characteristic of mountain snowfall 
in a belt of westerly winds. The variation of snowfall should be 
greater than the variability of total precipitation, because in the 
case of the former there are two variables, storminess and the 
critical temperature for snow formation. An example of the 
extreme variation which may be found is recorded at Lente in 
Savoy. This station had 1.81 meters of snow in 1921 and 6.03 
meters in 1922. A generalized table of the relation of snow dura- 
tion and altitude is that of the Harz Mountains of Central Ger- 
many. 



HUMIDITY AND PRECIPITATION 45 

Snow Duration in the Harz Mountains 

Altitude in meters 240 400 550 700 850 1000 1 150 

Duration in days 60 82 104 122 136 162 180 

The depth of snow cover as increasing with altitude is demon- 
strated by this Swiss table: * 

Average Annual Depth of Snow in Switzerland (1896- 1905) 

Depth of snow in 
Station Altitude in meters centimeters 

Elm 963 416 

Engleberg 1018 437 

Davos 1 56 1 519 

Sils-Maria 181 4 408 

Berhardin 2073 955 

St. Bernhard (1874-83) 2476 1050 

In the valley of the Inn at 2500 meters there are but 8 weeks 
free from snow on shady slopes. Permanent snow is found at 
3000 meters. In the Vallee de Conches at 1370 meters snow will 
reach the depth of three and more meters. In 1496 drifting snow 
covered the church at Minister. At 1700 meters the depth has 
been 7 meters. Tall poles must be set to guide travelers. 

One of the most decisive elements of mountain climatology is 
the snow line. By the snow line is meant the lower limit of snow. 
There are two types of snow lines. The climatic snow line is the 
highest of the lower limits of the snow, that which is seen in mid- 
summer. Though the general meteorological elements enter into 
this limitation, it may well be also the limit of neve or hardened 
snow. The second type of snow line is the orographic snow line. 
This is the lowest of the lower limits of snow, that of mid-winter. 
Snow lines as limits to human cultures should be considered by 
months. Perhaps the snow lines of April, May, and June are 
more critical than those of other months. Obviously, and espe- 
cially considering the preceding table of snow duration in Ba- 
varia, the snow line of spring is more significant than that of 
autumn. (Figure 10.) 

Fritzsch gives the following numerical relations between cli- 
mate and orographic snow lines. The regularity of the differences 
is worthy of note: 

1 Maurer. 



4 6 



MOUNTAIN GEOGRAPHY 



Snow lines in the Ortler Alps 



North East 

Climatic snow line .... 2870 m. 2940 m. 

Orographic snow line . . 2540 2640 

Difference 330 300 



South West Average 

3060 m. 2990 m. 2965 m. 

2750 2630 2640 

310 360 325 



The elevation of the snow line is the result of balance between 
supply and wastage. What is perhaps the highest climatic snow 
line in the world is in the Himalayas at 6100 meters, as reported 
by Strachey in the Encyclopaedia Britannica. Here is illustrated the 
question of supply and wastage. The extreme elevation is due to 
meager supply. The climatic snow line in the Alps varies locally 



% 

100 



30 



20 



. ^--.-—^ 

• / 

e -^—A 

f • 

< ►/__ 

/ • 

4 jT 

/ • 

--*£+- * 

/ • 

v —l ^ ! 1 

I I I I 1 1 1 1 1 1 1 1 1 \ 2 



U IN O Q E D S 



?* 24 ze ra JO 32 

rs/l tl T EI R 3 



Fig. 7. Percentage of Total Precipitation in Snow, Swiss Alps 
{.After Maurer.) 

The dots represent the stations of observation. The line, an average, shows a con- 
sistent increase of snow percentage with increasing altitude. 



HUMIDITY AND PRECIPITATION 47 

between 2700 and 3100 meters according to latitude, exposure, 
cloudiness (which affects evaporation), and mass of mountain. 
Where precipitation is slight and annual range of temperature is 
small the snow line will be low, that is, below the level of the 
mean annual isotherm of o degrees. The relation of the snow 
line to isotherm levels is discussed in another section. 

The height of the orographic snow line in Switzerland is indi- 
cated by the following very general table (see also Figure 11): 

Duration of Snow Cover According to Altitude in Switzerland 

Altitude in meters Number of days 

650 77 

1300 200 

195° 2 45 

If the snow line, in a general way, has a mean annual tempera- 
ture of o degrees there should be a decrease of elevation with 
increasing latitude. The failure of precipitation in the trade wind 
belts in continental stations causes a break in the curve. The 
following table was compiled by the late Robert DeCourcy Ward: 

Climatic Snow Line in Northern Hemisphere High Latitudes 



Latitudes 


Locality 




Alt. in meters 


82° 


N 


Franz Joseph's Land 




I OO—30O 


77 




Spitzbergen 




460 


73- 


74 


Nova Zcmbla 




60O 


70- 


71 


Norway: coast 
inland 




700-800 
IOOO 


70 




Greenland: inland 




700-800 


64-65 


Iceland: north side 




3OO 






south side 




60O 


60 




Mt. Saint Elias: west 


slope 


160O 






east 


slope 


180O 


55 




Kamchatka 




160O 


5° 




Vancouver Island 




I 60O- I 80O 


49 




Cascade Mountains 




2000 


47 




Tyrolean Alps 




2820 


45 




Mt. Hood 




2250 


43 

42 




Pyrenees: north side 
south side 
Mt. Shasta 




279O 
24OO 
24OO 



There is no climatic snow line for the northern tropical re- 
gions, though isolated patches of snow remain upon the moun- 
tains. The one exception to the increasing elevation in the table 



48 MOUNTAIN GEOGRAPHY 

is the low record of Mount Hood-, where the supply of snow is 
excessive; In the Arctic the climatic snow line is not known on 
land surfaces, but in the Antarctic it does reach sea level. Much 
geography may be read into the foregoing table. Latitude, 
marine influences, continentality, and exposure are indicated. 
Mass of mountain is hardly less important than the other factors. 
Snow lines are generally 600 meters lower in the central massifs of 
Switzerland than in the border ranges. In this respect mass of 
mountain can be thought of as influencing temperature. 



NW W 



5W 



5E 



NE 



CHmatfc 
Snowline 



Orographic 
Snowline 























^~ 


N 














y^ 














/ 


1 


\ 

\ 












/ 






\ 










/ 








^ 






> 


/ 








\. 






/ 










\ 

\ 






/ 
















/ 












\ 


S*' 








-—— 


















\ 










) 






\ 


\ 
















\ 








/ 
















/ 








\ 








/ 








x ^ 








/ 













































































* 



to 



cy 



U 



ui 



Fig. 10. Snow Limits in the Ortler Alps, Austria. {After Fritsch.) 



An obvious element in the distribution of snow is the degree of 
slope. A steep slope will not hold snow. Such steep eminences as 
K 2 and Nangi Parbet in the Himalayas shed snow in avalanches 
to such a degree that glaciers do not originate at the peaks but far 
below at a break in slope. 








■^s 



M 



Fig. 8. Snow Collkctinc I'n.i.ns of the Afftsch Glacier, Blrnksk Oberland 




•■^&\ 



1-. »*1F -*- 





FlG. 9. TlJNNEl. IN AN AVALANCHE, SWISS Al.PS 

A scene in June. It is frequently cheaper to drive a tunnel through an avalanche 
deposit than to clear a passage. This deposit is from an annually recurrent slide. 



HUMIDITY AND PRECIPITATION 49 

The best example of the influence of precipitation upon the 
snow line is found in transverse valleys in mountains sufficiently 
high to be true climatic barriers. The Himalayas are not every- 
where exposed to the moisture bearing winds. At Leh in Western 
Tibet a snow cover of 3 centimeters would be normal and in the 
high passes at 5100 to 5800 meters snow cover is frequently not 
more than 6 centimeters. Yet eastward, where the ranges are 
struck by monsoon winds, the passes are blocked from September 
to mid-June. 

Of the two controls, precipitation and solar heat, we may do 
well to regard precipitation as the greater in determining snow 
lines. Again referring to the Himalayas, between 27 and 34 de- 
grees North latitude we find that on the Indian or wet side the 
average climatic snow line is 4900 meters, while on the Tibetan 
or dry side the line is 5600 meters. Supply here negatives the 
influence of exposure to sun. Moisture and aridity seasons often 
counteract the temperature seasons. In the mountains along the 
Upper Mekong the winter snow line is 5180 meters, while in 
summer it is only 4270 to 4570 meters. This is because of the 
excessive aridity of the monsoon winter. As latitude increases, 
the matter of exposure to sun becomes relatively more important. 
This is illustrated by the south and north exposures in the Inntal 
of Austria. 

Exposure and Snow in the Inntal 

South slope North slope 

March 960 720 

April 1 2 70 1 1 1 o 

May 1 700 1 540 

June 2 1 90 2030 

July 2680 2470 

August 3130 2930 

September 3210 2760 

October 2 1 50 1 890 

November 1300 1010 

December 740 680 

The Caucasus illustrates at once the factors of precipitation 
and temperature. 

Mean Climatic Snow Lines of the Caucasus 

Latitude North slope South slope 

Western ranges 47«7~43-5 N 3400 2920 

Central ranges 41.5-42.7 3300 3230 

Eastern ranges 40.5-41.5 3600 3720 



50 MOUNTAIN GEOGRAPHY 

As the Peruvian Andes are north-south ranges in low latitudes 
the question of temperature exposure is largely eliminated and 
precipitation exposure is demonstrated. Near Quito on the 
snowy exposure the line has a mean altitude of 4560 meters while 
the dry side has an altitude of 4740 meters. 

Human Responses to Snow 

Though the human responses of climate are for the most part 
assigned to later chapters, it is convenient to consider here the 
importance of snow and the importance of avalanches. Snow has 
four types of influences on human affairs: 

1 . Direct influence of snow cover. 

2. Influence of snow on temperature. 

3. As a source of water supply and floods. 

4. As avalanches. 

One of the best studies of the influence of snow upon the life of 
mountain folk is that of Aimee Bigallet, a school teacher in the 
hamlet of Etages, in the commune of Saint-Christophe-en- 
Oisans. In the winter recorded, that of 1921-22, the hamlet 
of Etages had 160 centimeters of snow in January alone. The 
mountain torrent roared under a roof of ice. Wind shook the 
houses until the furniture trembled. February saw 69 centi- 
meters of snow, and even March had 14 days of snowfall. Yet the 
altitude was great enough to permit a degree of insolation, so that 
in spite of low air temperatures the school children ate out of 
doors on March 17th, and the men gathered for gossip on the 
only level spot in the hamlet, the bridge across the stream. Com- 
munications were broken ; the telephone was down on November 
1 8th and repaired in June. The post got through only between 
storms and when there was not too great danger of avalanches. 
There was no beast travel. All movement outside the village 
was on skis. A child fell ill in March; the father was not able to 
get down the valley for a physician's advice and remedies until 
April 1 2th; the child itself could not be taken to the doctor until 
a month later. A dead body was retained five days in February 
before a party of men on skis could be gathered to remove it. 
Dead bodies in some valleys are frozen and kept for months. 

Descriptions of the influence of isolation due to snow upon 
cottage industry are numerous. Men who would otherwise be 



HUMIDITY AND PRECIPITATION 51 

idle have time to carve wooden toys, to construct clocks and 
watches carefully, to manufacture optical glass and scientific 
instruments. The women make lace and do weaving and em- 
broidery work. Though water-power or labor supply has cen- 
tralized these industries in such places as Geneva and Ncuchatel, 
the delights of summer life on a Swiss farm will always keep 
from the city many people who in winter will turn to the 
cottage industry. 

In winter the Swiss live in substantial and often ornate houses 
of wood and stone, since both materials are at hand. Because the 
buildings are low, they are more easily heated. Where there is 
much snow, the roofs are usually high-pitched so that the snow 
may be easily shed. However, in the Alps there are places where 
the winds are so strong that, to withstand them better, the roof is 
low and the slabs or shingles are weighted down with stones. The 
period of indoor living is long, since the snow may lie upon the 
ground for seven or eight months of the year. During this period 
poor ventilation and lack of exercise or mental stimulus have a 
decided ill effect upon health. This condition, plus the hard 
struggle for existence, may be a reason for the large number of 
insane in the more remote districts. The tavern life of the little 
villages is another unfortunate result of the indoor confinement. 
Certainly the rigor of winter life is among the causes of the large 
seasonal migration. 

The long duration of the snow cover limits the growing season. 
No matter how high are the temperatures of air and soil on 
cleared areas, the regions given over to snow arc more or less 
dormant. Even trees are limited in their upward expansion by 
snow, as in the Tirol at 2400 meters, where the duration of snow 
is nine and one-half months. On the other hand, a moderate 
duration and depth of snow cover is beneficial to both trees and 
grain, because it protects against severe cold. To quicken melting 
in spring, earth is at times scattered over the snow. 

Snow greatly facilitates transportation. This is true of valley 
roads where there is flattish land and not too great danger from 
avalanches. Little trails on steep slopes are difficult under 
snow. Logging on slopes is greatly facilitated by winter. The 
common custom in European mountain meadows of building the 
hay barn in the center of the meadow instead of near the stable, 
indeed, it may be, several miles from the stable, is explained by 
the fact that hay is more easily transported in winter over the 



52 



MOUNTAIN GEOGRAPHY 



snow than over summer roads. More often than not, the barn is 
up the mountain from the stable. The hay is brought down in 
winter on sleds, when the difficulty is not so much in moving the 
hay as in controlling the speed at which it descends to the valley. 
Clear cold days and nights may witness an almost constant pro- 
cession of sleds coming to the village. So isolated are many of the 
meadows that without snow transportation the hay could be 
brought down only with the greatest difficulty. In mountains 
where snow is scanty, hay may be brought from the meadows by 
aerial cables. 



*o 



ft 



Id 



id 



^SOC 


J 


F 


M 


A 


M 


J 


J 


A 


S 





N 


D 


COCO 
3500 


























zooo 
1500 












Snow- Free 




















Period 










/ooo 
























« 



Fig. 1 1 . Snow Free Period in the Oisans According to Altitude 
(After Blache.) 



The relation between snow and temperature readily suggests 
itself. By reducing mountain temperatures to sea level one has 
evidence of the influence of snow on temperature. Benevent has 
done this for the French Alps. He finds Ventoux in the Southern 
Alps a cold island in winter because of snow. The degree to 
which snow cools air may be suggested by the differences in snow 
and air temperatures. 

A characteristic of high lands in spring is that the season will be 
well advanced before the snow disappears. Once the snow is 



HUMIDITY AND PRECIPITATION 



53 



gone the soil, easily dried because of its slope position, warms 
quickly. The rapidity of budding and flower growth is one of the 
delights of spring at high altitudes. Because the snow remains in 
patches well into the warm season, flowers push up through the 
snow. During the period when snow has disappeared from the 
valleys but not from the uplands there is, naturally, an increase 
of the vertical temperature gradient over the normal. 

Air and Snow Temperatures at Davos 
(1560 meters alt.) 

7 A.M. 



Days of Observation 

1 89 1 February 
March 
December 1 

1892 January 
February 1 



28 

3 1 

12 

3 1 
20 



Air 

- 12.0° 
~ 5-3 

- 9-3 

- 9-7 

- 6.0 



Snow 
-I6. 7 ° 

- 7-o 
-13-9 
-1 1.9 

- 7-5 



Difference Amt. of cloud 



4-7 
i-7 
4.6 
2.2 
i-5 



2.2 
6.6 

4-5 
6.2 
6.6 



122 


Mean- 


- 8.5 

2 P.M. 


- 1 1 .4 


2-9 


Days of Observation 




Air 


Snow 


Difference Amt of c 


1891 February 28 
March 3 1 
December l 1 2 




i-9° 
2-9 
-3- 1 


- 7-2° 

— I O.I 


9.1 1.8 
4.0 0.4 
7.0 4.0 


1892 January 31 
February x 20 




-i-7 
1.2 


- 6.0 

- 3-2 


4-3 5-6 

4.4 6.6 



122 



Mean 0.2 



~ 5-5 



5-7 



The snow of high elevations melts slowly. It is warm air tem- 
perature rather than direct insolation that is effective in melting 
snow. Snow reflects solar heat some ten times more than does 
soil. This is true because of the lack of color and the crystal sur- 
face of snow. Therefore insolation values during the snow cover 
are increased by the amount of reflected light and heat. 

The most recent development of mountain studies is that of 
snow surveying. All work along this line is of the present century 
and most of it of the last decades. The object of such studies is to 
ascertain the amount of snow stored on the heights, in order to 
forecast the possibilities of irrigation and water supply for the 
summer, and, secondarily, to indicate the probability of floods. 

Snow storage depends upon not only the amount of precipita- 



Records not complete. 



54 MOUNTAIN GEOGRAPHY 

tion and the extent and slope of terrain but also the weather con- 
ditions during the storage period. Moreover, depth of snow is not 
necessarily a measure of the water content of a field. Snow has 
compressibility. The difficulties of judging the water content 
have been more nearly overcome by American students than 
by the Europeans. The article of J. E. Church on "Snow Sur- 
veying" meets the need of students in that subject. His article 
goes into conservation of snow, forecasting of run-off, floods, and 
certain matters not discussed in this volume. Some generalities 
and statements from Church's article are given here: 

The higher the mountains the longer the period of snow storage. 

The evaporation of snow on mountain heights is great. On the 
summit of Mt. Rose (10,800 feet) in 69 hours an evaporation of 2.32 
inches, water content, was recorded. 

Evaporation will occur at night and below freezing temperatures. 
A 3 1 -mile wind created an evaporation of snow of .08 to .10 inches in 
a single night. 

The efficiency of mountain forests as snow gatherers and conservation 
is shown by contrasts between measurements on the summit of Mt. 
Rose and forest stations in April, 1910: 

On summit: 

Sheltered by Observatory 52.5 inches 

Wind-swept slope 8.1 " 

Protected slope 78.1 " 

Average of unforested talus slope 40.8 " 

Average forest station 88.6 " 

Talus slopes are less efficient conservers of snow than forest slopes. 
Dense forests hold snow on trees and increase evaporation. 

The depth and character of the cover is a basis of forecasting 
available water for irrigation. So accurate are the methods that 
farmers have changed their crop plans because of the forecast. 

A. J. Henry has studied the relations between snow and floods. 
He lays stress on the unimportance of insolation as a factor in 
snow melting and as a cause of floods. Flood waters from snows 
come only at a time of high night temperature. Fresh snow melts 
more rapidly than old, packed snow. The decreasing depth of 
snow is not always a measure of the melting. Depth of snow may 
decrease by (1) settling, (2) packing by wind, (3) absorption by 
ground when not frozen, (4) evaporation. 

Lastly, snow as a source of water may pass through the ice 



HUMIDITY AND PRECIPITATION 55 

stage. Glaciers melt slowly. Curiously, several of the large 
American cities depend on water which originates with glacial 
streams. As the glaciers are becoming seriously reduced in size, 
there is genuine concern as to the water supply of the future. 

Snow Slides 

The varied forms derived from the Latin word for slide, labina, 
suggest the ubiquity of slides in mountain regions. We have for 
snow slides the terms lavina, lavigna, avalanga, levina, avalanche, and 
valanga. The significance of the place names Lavanchy, La- 
vancher, Lanch, Lanchettes, and Lakne is apparent. 

An obvious factor in the retention of snow on steep slopes is the 
character of the surface upon which the snow lies. Is the ter- 
rain a continuously downward slope, or is it rough and uneven 
in its grade? Are there rocks, ledges, or fences to hold the snow? 
Is there a bush or forest covering? The slope necessary for slides 
depends upon weather conditions, the quality of the snow, and 
the topography. A very general set of slope limits for snow with- 
out slides is: 

22 degrees slope and 40 to 50 centimeters of snow 
50 " " " 5 " " " 

Allix is our chief student of the avalanche. He divides snow- 
slides into two classes: the cold avalanche and the warm ava- 
lanche. The cold avalanche involves the movement of dry snow. 
It takes place in mid-winter at the time of greatest cold and often 
after a drop in temperature. Though the snow involved in a 
slide may lie directly upon the ground, movement is more likely 
if the snow lies upon an icy surface developed in the autumn 
and subsequently covered with the dry snow. A chief danger of 
the cold avalanche is the impacting of the snow as it strikes its 
resting place. This type of avalanche is perhaps the equivalent 
of the Staublawine of the Germans. This is a dust-snow avalanche 
attended by a whirlwind. This cold snow dust may also slide in a 
less catastrophic manner. In a ravine dry snow may descend as a 
powder 'waterfall' for days. The dust avalanche which com- 
mences upon high alps and gains the oversteep slope of the lower 
valley may roar toward the valley way at 100 to 200 kilometers 
an hour. 



56 MOUNTAIN GEOGRAPHY 

The warm avalanche (warm is here a relative term) is of damp 
snow. Such avalanches are water-soaked, heavy, and tend to act 
as a single mass. Their destructiveness can be easily imagined. 
The descent may be as a cascade, bringing destruction unwarned 
to a village. More often the descent is as a landslip, a relatively 
slow movement but one of force to be feared. Such a snow slip 
will have a velocity of from a few to 50 miles per hour. This 
warm avalanche is the possible equivalent of the German Schlag- 
lawine. In the period including the winters of 1900—01 and 
1 91 3-1 4 in Savoy, of 586 avalanches, 421 were warm ava- 
lanches, 31 were cold avalanches, and two were glacial. 

Protective works can be divided into two classes, preventive 
and curative. 

(1) Preventive methods aim at precluding the formation of 
the avalanche by retaining the snow on the high slopes. This has 
been done in many cases by the construction of wooden or stone 
fences. There should be reforestation of high slopes wherever 
feasible. Avalanches may start from above or below the tree 
line. About two-thirds of the ground avalanches and about half 
of the other avalanches start below tree line. This indicates the 
importance of reforestation. Stone walls of a meter or more in 
height and a few meters long are scattered over steep hillsides. 
Slat fences arc used in the same manner. In the Lower Engadine 
ditches one meter broad, of similar depth, and six to nine meters 
long, were built as protection against avalanches. Today pro- 
tection against the avalanche is so common a feature of Alpine 
valleys as to excite little curiosity in the traveler. (See Figures 
12 and 13.) 

(2) Curative methods consist in devices to restrict the destruc- 
tive power of avalanches and include the building of strong 
masonry walls in the shape of plowshares. In Switzerland arti- 
ficial channels in stonework have been built to carry off the tor- 
rent of snow. Roads and railroads which arc forced to cross the 
path of avalanches may have a stone shed built over the road 
with a slanting roof. The slide then passes over the roadway. 
When roads are not so protected and the snow blocks the road, a 
tunnel may be driven through the snow. The tunnel may last 
far into the summer. (Figure 9.) 

There is little possibility of forecasting the precise moment of 
release of an avalanche. The only observers qualified to speak on 
this matter have all been impressed by the suddenness of the 




Fig. 12. Piers to Protect Poles against Avalanches, Engadine 
(Courtesy of the Rhaetischebahn.) 



, P 




Fig. 13. Fences Used to Prevent Avalanches 
(Courtesy of the Rhaetischebahn.) 



HUMIDITY AND PRECIPITATION 57 

occurrence. One's best guide in escaping the dangers of ava- 
lanches is prudence. Trifling causes such as the following may 
start an avalanche: the passage of a party of persons; a slight 
shaking of the ground; the sounding of a distant train; or bells of 
the flocks. There is reported to be a local regulation against the 
Swiss national habit of yodeling, lest snow slides be started. 
There are critical periods during which avalanches are particu- 
larly to be feared. At such times the snow is in a position of 
unstable equilibrium. 

In the prediction of avalanches the most one can hope to obtain 
is the determination of this state of unstable equilibrium, indicat- 
ing the liability to accidents at critical periods. The only prac- 
tical method, outside of native intelligence, is the preparation of 
meteorological probability charts. For this it would be necessary 
to determine for a number of years, by means of daily observa- 
tions taken in the high mountain regions, the meteorological 
characteristics of the periods that precede the starting of ava- 
lanches. This would include pressure, winds, temperature, and 
precipitation. In addition, one should know the density and 
temperature of the snow and the condition of the surface upon 
which the snow rests. As has been said, an icy surface developed 
in the autumn will increase the probability of avalanches 
through the winter. A marked rise in temperature, an accumula- 
tion of soft snow, or a wind striking the inclined snow fields are 
conditions which indicate probability of an avalanche. 

In high mountains whose slopes have been oversteepened by 
glaciation avalanches are the rule rather than the exception. It 
is difficult for a plainsman to appreciate the extent of the danger. 
In Switzerland there are 9368 avalanche corridors, of which 
5294 see service many times a year. There are, in addition, many 
avalanches which occur occasionally but do not develop regular 
corridors. A limited region in the French Alps, the Tarentaise, is 
known to have 46 important slide corridors which are greatly 
feared for their recurrent avalanches. 

The erosive effect of the avalanche on the mountain surface, 
aside from the damage done in the zone of habitation, is a matter 
of no small moment. It is true that the work of an avalanche is 
over in a moment, but during this moment it can move more 
earth than the stream in the ravine during the entire year. 
Mougin measured the amount of material dropped by avalanches 
in the inhabited valleys of Savoy from 1908 to 191 2; the total 



58 MOUNTAIN GEOGRAPHY 

volume of earthy material thus carried amounted to 44,000 cubic 
meters. If we assume that a large ground avalanche descends 
through 2000 meters with a speed of 10 meters a second and 
a mass of 200,000 cubic meters, its energy is approximately 
20,000,000 horse power. 

A second danger is the destruction of forests. The ravines 
scoured each year by avalanches are denuded of trees and new 
tree growth there is impossible. Moreover, there is a pushed 
wind which precedes the avalanche. The writer has seen a tree 
two feet in diameter which was broken off 20 feet from the 
ground by such a wind in the Yosemite Valley. This blast may 
create havoc with a forest before the snow actually strikes. The 
blast not infrequently is forced upon a forest of an opposite slope 
so as to destroy forests beyond the reach of the snow. An example 
of this is to be found in a valley above Pralognan, French Alps. 
Reports state that avalanche winds are felt kilometers away from 
the snow slide. 

Some minor results of avalanches are the favorable deposition 
of soil in valleys, the early removal of snow from high level fields, 
the provision of wood for fuel at the base of the slide, the storage 
of water in the compact and slowly melting snow deposit of the 
valley. There is also a local lowering of temperatures from the 
accumulation in the valley. 

Avalanches in certain valleys prevent winter travel. So fre- 
quent are they that they form a constant danger to the traveler. 
It is the slides rather than the depth of snow cover which isolate 
certain villages. This is particularly true because, while some- 
times they follow known tracks, at other times they are as ca- 
pricious as lightning. In addition to the element of danger, the 
slides often obstruct roads. Travel may temporarily go over the 
deposit, roads may be dug out, or, as previously mentioned, tun- 
nels may be constructed. Telephone and telegraph lines, of 
course, suffer. The villagers of St. -Maurice and St. -Jacques in 
the French Alps are held virtually prisoners for months by the 
fear of avalanches. The Rhone Valley of Switzerland has 
numerous crosses marking the place of death of winter travelers. 

In 1 879, along the 5-kilometer route from Toesch to Zermatt, 
there was no space of 50 meters free from avalanche deposit. In 
the same year in the Saas and Binnen valleys it was dangerous to 
venture from the houses. In 191 7 one valley in Switzerland 
counted 30 avalanches. That year there was great loss of life, 



HUMIDITY AND PRECIPITATION 59 

houses, barns, and cattle throughout the Alps. A railway train 
at Wolfgang near Davos was destroyed with the loss of ten lives. 
The country gave 192,000 francs for relief in the devastated 
areas. Swiss history is indeed full of snow slide catastrophes. On 
January 16th and 17th of the year 1594, slides tore down at Be- 
dretto in Switzerland a church, a rectory, houses, and barns. 
On January 22nd, 1634, the priest's house was destroyed and the 
priest himself buried. In 1695 the snow measured there 3.6 me- 
ters and slides came from all sides. The church, houses, and 
barns were destroyed. 

In the valley of St. Antonien in Grisons from 1608 to 1876 
there were destroyed 38 houses, 200 barns, 4 saw mills, 43 per- 
sons, and 130 head of cattle. In 161 8 at Henkerbad 61 persons 
were killed. An avalanche in Valais in 1 719 wiped out a town, 
killing 60 persons; that of Obergestelen in 1720 destroyed 120 
houses and killed 80 persons and 400 head of cattle; that of Saas 
in Grisons in 1689 wiped out the village and all the inhabitants. 

On the 20th of February, 1720, the Swiss village of Ober- 
gestelen was destroyed and 84 of its 200 inhabitants were killed 
by an avalanche which leaped an intervening forest and wrecked 
one-third of the village. In a dairy country the loss of a cow is a 
calamity, and at that time six hundred head of cattle were de- 
stroyed. The snow from the avalanche blocked the river Rhone, 
which quickly cut through it and then flooded a portion of the 
settlement. The remainder of the town was laid in ruins by a 
fire. The indomitable villagers rebuilt their town, only to have 
it subsequently destroyed by an avalanche from the other side of 
the valley. 

BIBLIOGRAPHICAL NOTES 

Evaporation 

Camp, W. H. "The Grass Balds of the Great Smoky Mountains of 

Tennessee and North Carolina," in Ohio Journal of Science, xxxi 

0930, PP. 157-164. 
Chodat, Fernand. "Resultats d'une enquete atmom6trique au jardin 

alpin 'La Linnaea,'" in Archives des sciences physiques et naturelles, 5 m0 

periode, ix (1927), supplement, pp. 126-127. Evaporation rates of 

Alpine habitats. 
Gates, F. C. "Evaporation in Vegetation at Different Heights," in 

American Journal of Botany, xiii (1926), pp. 167-178. Also, in part, in 

Monthly Weather Review, liv (1926), p. 61. 



60 MOUNTAIN GEOGRAPHY 

Harshberger, J. W. " Alpine Fell-Fields of Eastern North America," in 

Geographical Review, vii (191 9), pp. 233-255. 
Livingston, B. E. "Evaporation and Plant Habitats," in The Plant 

World, xi (1908), pp. 1-9. 
Liidi, Werner. "Die Ergebnisse von Verdunstungsmessungen in 

Lauterbrunnental und in Bern in den Jahren 191 7 bis 1920," in 

Festschrift Carl Schrbter (Zurich, 1925), pp. 185-204. Good. 
Maurer, J. "Verdunstungsmessungen an Freien Wasserflachen in 

Hochgebirge," in Verhandlungen der Klimatologischen Tagung in Davos, 

J 9 2 5> PP- 1 19-126. 

Precipitation 

Ahlmann, H. W. "Precipitation Conditions in the Horung Massif," in 

Geografiska Annaler, ix (1927), pp. 21-35. 
Benevent, Ernest. " La pluviosite de la France du Sud-Est," in Recueil 

des travaux de VInstitut de geographic alpine, i (191 3), pp. 323-442. 
Benevent, Ernest. "Du role des montagnes dans la formation de la 

pluie," in Revue de geographic alpine, xii (1924), pp. 173-187. 
Blanchard, Raoul. " Enormes precipitations dans le massif de la 

Chartreuse," in Recueil des travaux VInstitut de geographic alpine, vii 

(1919), pp. 717-720. 
Brockmann-Jerosch, Heinrich. Die Niederschlagsverhaltnisse der Schweiz* 

Zurich, 1925. 
Brooks, C. F. " Orographic Rainfall," in Geographical Review, xv (1925), 

p. 31 1. A review of Benevent's article of 1924. 
Goodnough, X. H. "Rainfall in New England," in Journal of the New 

England Water Works Association, xxix (19 15), pp. 237-438. 
Hann, Julius. "Der Regenfall auf den Hawaii-Inseln," in Meteorolo- 

gische ^eitschrift, xii (1895), pp. 1— 14. 
Hellmann, Gustav, and others. Klitna- Atlas von Deutschland. Berlin, 

1 921. 
Henry, A. J. "Increase of Precipitation with Altitude," in Monthly 

Weather Review, xlvii (19 19), pp. 33-41. 
Huber, Rudolf. Die Niederschlage in Kanton Basel in ihrer Beziehung zu den 

orographischen Verhaltnissen. Zurich, 1894. 
Jefferson, Mark. "The Rainfall of Java," in Geographical Review, v 

(1918), pp. 492-495- 
Knoch, Karl, and Reichel, Eberhard. Verteilung und J a hr lie her Gang der 

Niederschlage in den Alpen. Berlin, 1930. (Veroffentlichungen des Preus- 

sischen Meteor ologischen Instituts, Nr. 375.) Important. 
Lee, C. H. "Precipitation and Altitude in the Sierra," in Monthly 

Weather Review, xxxix (191 1), pp. 1092-1099. 
Lee, W. T. "The Raton Mesas of New Mexico and Colorado," in 

Geographical Review, xi (1921), pp. 384-397. Human significance. 



HUMIDITY AND PRECIPITATION 6 1 

Liitschg, Otto. Uber Niederschlag und Abfluss im Hochgebirge. Zurich, 
1926. (Schweizerischer Wasserwirtschaftsverband, Nr. 14.) 

McAdie, A. G. "The Rainfall of California," in University of California 
Publications in Geography, i (191 4), pp. 127-240. Excellent. 

Varney, B. M. " Monthly Variations of the Precipitation- Altitude Rela- 
tion in the Central Sierra Nevada of California," in Monthly 
Weather Review, xlviii (1920), pp. 648-650. 

Wells, E. L. " Precipitation in Oregon," in Monthly Weather Review, 1 
(1922), pp. 405-411. 

Snow 

Allix, Andre. "Neiges d'ete en 1922 ct leur l limite instantanee,' " in 
Revue de geographie alpine, x (1922), pp. 649-664. 

Benevent, Ernest. " La neige dans les Alpes franchises," in Recueil des 
travaux de VInstitut de geographie alpine, v (191 7), pp. 403-498. 

Bigallet, Aimee. "La vie d'hiver dans le Haut-Veneon," in Revue de 
geographie alpine, x (1922), pp. 625-634. Excellent. 

Brooks, C. F. "New England Snowfall," in Monthly Weather Review, 
xlv (191 7), pp. 271-285. Heavy snow on low mountains. 

Bruckner, Eduard. "Die Eiszeiten in den Alpen," in Geographische 
Zeitschrift, x (1904), pp. 569~57 8 - 

Bruckner, Eduard. " Uber den Einfluss dcr Schneedecke auf das Klima 
dcr Alpen," in ^eitschrift des Deutschen und Osterreichischen Alpen- 
vereins, xxiv (1893), pp. 21-51. 

Church, J. E. " The Conservation of Snow: Its Dependence on Forests 
and Mountains," in Scientific American Supplement, lxxiv (191 2), pp. 
152-155, and, in German translation, in Meteor ologische Zjeitschrift, 
xxx (1913), pp. 1-10. 

Church, J. E. "Snow Surveying," in Geographical Review, xxiii (1933), 
pp. 529-563. An important article. 

Dainelli, Giotto. "II limite delle ncvi nel bacino superiore dell' Indo 
(Caracorum e Himalaia occidentalc)," in Recueil de travaux ojfert a M. 
Jovan Cvijic (Belgrade, 1924), pp. 1-10. 

Dansey, R. P. "The Glacial Snow of Ben Nevis," in Symons's Meteoro- 
logical Magazine, xl (1905), pp. 29, 32. 

Ficker, H. von. "Der Einfluss orographischer Verhaltnisse auf die 
Niederschlage in Tirol und Vorarlberg," in Meteor ologische Zeit- 
schrift, xxvi (1909), pp. 31 1-3 1 5. 

Fisher, L. C. "Snowfall on Mount Rainier," in Monthly Weather Re- 
view, xlvi (1918), pp. 327-330. 

Henry, A.J. " The Disappearance of Snow in the High Sierra Nevada 
of California," in Monthly Weather Review, xliv (191 6), pp. 150-153. 

Hofmeier, Walter. " Die Hochalpen im Winter," in Z e ^ tscnr if t des Deut- 
schen und Osterreichischen Alpenvereins, lv (1924), pp. 246-268. 



62 MOUNTAIN GEOGRAPHY 

Huber, Rudolf. Die Niederschldge in Kanton Basel in ihrer Beziehung zu den 

orographischen Verhaltnissen. Zurich, 1894. 
Jegerlehner, J. "Die Schneegrenze in den Gletschergebieten der 

Schweiz," in Beitrdge zur geophysik, v (1903), pp. 486-568. A factual 

article. 
Knoch, Karl, and Reichel, Eberhard. Verteilung und jdhrlicher Gang der 

Niederschldge in den Alpen. Berlin, 1930. (Verqffentlichungen des Preus- 

sischen Meteor ologischen Instituts, Nr. 375.) 
Koppen, Wladimir. "Die Lufttemperatur an der Schneegrenze," in 

Petermanns Mitteilungen, lxvi (1920), pp. 78-80. 
LeConte, J. N. " Snowfall in the Sierra Nevada," in Sierra Club Bulletin, 

vi ( 1 907-08) , pp . 3 1 0-3 1 4. A famous American student of mountains. 
Maige-Lefournier, Mathilde. " La vie traditionnelle hivernale a 

Bonneval-sur-Arc (1846 m.)," in La montagne, xiii (1917), pp. 1-16. 

Good description. 
Maurer, J. "Temporare Schneegrenze und mittlere Schmelzwasser- 

hohen im schweizerischen Alpengebiet nach neueren Erhebungen," 

in Meteor ologische ^eitschrift, xxvi (1909), pp. 539-546. 
Palmer, A. H. " The Region of Greatest Snowfall in the United States," 

in Monthly Weather Review, xliii (191 5), pp. 217-221. 
Ratzel, Friedrich. "Die Schneedecke, besonders in deutschen Ge- 

birgen," in Forschungen zur deutschen Landes- und Volkskunde, iv (1890), 

pp. 107-277. 
Sherier, J. M. " Mountain Snowfall and Flood Crests in the Colorado," 

in Monthly Weather Review, li (1923), pp. 639-641. 

Avalanches 
See especially Benevent under General References on Climate. 
Allix, Andre. "Avalanches," in Geographical Review, xiv (1924), pp. 

519-560. Repeated, with slight modifications, in the next item. 
Allix, Andre. " Les avalanches," in Revue de geographic alpine, xiii (1925), 

pp. 359-423. With a large bibliography. 
Alter, J. C. "Avalanche at Bingham, Utah," in Monthly Weather Re- 
view, liv (1926), pp. 60-61. 
Coaz, Johannes. Statistik und Verbau der Lawinen in den Schweizeralpen. 

Bern, 1910. 
Freeh, Fritz. " Uber die Lawinen der Alpen," in Pelermanns Mit- 

teilungen, lviii, 1 (191 2), pp. 126-128. 
Lunn, Arnold. "Mountaineering on Ski," in Mountain Craft, ed. by 

C. W. Young (London, 1920), pp. 397-470. 
Pollack, Vincenz. Uber die Lawinen Osterreichs und der Schweiz und der en 

Verbauungen. Vienna, 1891. 
Pollack, Vincenz. Uber Erfahrungen im Lawinenverbau in Osterreich. 

Vienna, 1906. 
Zdarsky, Matthias. Beitrdge zur Lawinenkunde. Vienna, 1930. 



CHAPTER III 
WINDS, CLOUDS, AND SUN 

Mountains and Winds 

MOUNTAIN areas, like plains, are subjected to prevailing 
winds. There are, however, a number of wind conditions 
which are characteristic of mountains. 

Mountains thrust their peaks into upper reaches where wind 
velocities are greater. 

Mountains form wind shelters. 

Mountains deflect winds and cause special local prevailing 
wind directions. 

Local valley relief may increase velocities. 

Out of physical conditions inherent to the mountain area there 
originate special winds. 

Exposure to prevailing winds means greater precipitation, or 
greater evaporation, or greater warmth or greater cold, as the 
case may be. These matters need no explanation. But reference 
must be made to the potency of these conditions. Even low 
mountain peaks suffer excessive evaporation from high wind 
velocities, more so apparently than equal altitudes on a flank of a 
large mountain. 

Wind velocity in free air increases with altitude. Unless 
broken relief of high areas is of such a nature as to offer consider- 
able friction, wind velocities increase with altitude among moun- 
tains. The rate at which winds increase in velocity with altitude 
is suggested by de Martonne in the following table: 

Wind Velocity and Altitude 

Mountain Altitude Relative velocity 

Krememunster 390 m. 3.5 

Santis Q500 7.5 

Sonnblick 3100 9.3 

Indeed, wind velocities may be greater near and about a peak 
than in free air at the same altitude. The reason for this is the 
same as that which causes stream currents to become stronger 
around an impeding boulder than in free channels. A wind of 



64 MOUNTAIN GEOGRAPHY 

164 miles per hour (262 kilometers per hour) has been measured 
in the winter of 1933-34 at the observatory on the summit of 
Mount Washington, New Hampshire. 

Sheltered valleys may be relatively calm, as tourists in the Swiss 
area can testify. Many of the winds there experienced are in- 
draft or reverse winds which are of slight velocity as compared 
with the prevailing winds. Peaks, on the other hand, arc seldom 
calm. In a year's observations valley stations as Allsaltcn in the 
Rhine Valley had 859 calms, Lugano 949, and Locarno 1001. 
The summit of Santis had but 72 calms and the Great St. Bernard 
but one calm. 

Even on plains lower winds do not always have the same direc- 
tion as that of the prevailing winds over the area at the moment. 
If a mountain range forms an angle with the prevailing wind 
direction there will be local deflections of surface winds as com- 
pared with the wind directions among and above the peaks. 
(Figure 14.) An excellent example of valley deflection of wind 
in the Davos valley, Switzerland, is given by Hugo Bach, Das 
Klima von Davos (Zurich, 1907), p. 86. 

Valleys transverse to the wind directions may have relative 
calm. Generally an obstacle mountain range protects a plane 
level for a length 16 times its height. This will vary slightly ac- 
cording to the wind velocity and the slope of the mountain. Such 
is, however, a theoretical rather than an actual protection. 
Winds may descend sharply on leeward sides of mountains. 
Mentone on the Riviera is protected from a strong north wind 
(the Mistral) by the Provencal Alps. But this Mistral may 
abruptly strike the sea a kilometer from shore. 

When a wind has a dominant trend coinciding with a valley 
axis, the velocity of the wind is intensified along the valley. Sc 
intense does the wind become in the valley of the Grand-Buech 
in the French Alps that it prevents progress in walking. At such 
times wind velocity in free air may not be intense and the side 
valleys may have comparative calm. 

Throughout the French Alps there is an Italian wind known as 
La Lombarde which has great force. The explanation is that 
when a cyclone lies over Southern France and anti-cyclonic con- 
ditions prevail over Italy there is difficulty in adjustment of pres- 
sures because of the mountain barrier. The wind pouring over 
the crest is particularly severe. In the Haute Tarentaise it is 
known as the 'Petit St. Bernard.' In 1882 La Lombarde de- 



WINDS, CLOUDS, AND SUN 65 

stroyed 52,000 trees near Chamonix and in 1904 it destroyed 
2600 cubic meters of wood (700 cords) in a forest on the French 
slope. At Seez large stones are needed to weigh down roofs. If 
the valleys of the French Alps were longitudinal this wind would 
evolve into a Foehn wind. The Foehn wind is described later in 
this chapter. 

Clouds in Mountains 

In addition to the irregularity of clouds according to the varia- 
tions of current weather, mountains impose upon prevailing 
winds certain special conditions and create special clouds. The 
result is that no weather element has greater irregularity than 
cloudiness. 




Fig. 14. Wind Roses for Switzerland 

Topographic controls are clearly indicated. The massive wind rose in the north- 
east corner is for the isolated peak of Santis and so represents winds of the free air. 
{After Frith.) 

There cannot even be declared a relationship between altitude 
and clouds. True, in summer the cold peaks condense moisture 
and so have hood clouds. The cloud zone of winter is lower than 



66 MOUNTAIN GEOGRAPHY 

that of summer. Hence the valleys, already deprived of much 
direct sunlight, are furthermore cloud-enshrouded, whereas the 
higher elevations are in sunlight. Also the nebulosity of valleys in 
winter is increased by local fogs. If one were to attempt a gen- 
eralization as to the relation between altitude and cloudiness, the 
statement might read in this fashion: Cloudiness increases with 
altitude within limits. These limits are those of stratus clouds 
and vary with season, latitude, exposure to wind, and degree of 
slope. It may be possible that the steeper the slope the greater 
the cloudiness. 

Hann took five principal cities on the Swiss plain, two towns in 
low valleys in Tirol, three high valley stations, and four pass 
stations. These he compared with observations on Sonnblick. 
His table shows the seasonal march of cloudiness at these com- 
posite elevations: 

Seasonal Cloudiness and Altitude 

(io equals overcast) 
Station Alt. in meters Winter Spring Summer 

Swiss plain 420 7.3 5.8 5.2 

Low valleys .... 1300 4.6 5.8 5.4 

High valleys ... . 1830 3.7 4.6 5.0 

Passes 2600 4.6 6.1 5.6 

Sonnblick 3100 5.2 7.1 7.3 

Cloudiness and, by the same token, sunshine are basic in an 
economic study of mountain life. Cloudiness is not numerically 
the complement of sunshine, as sunshine is partly a matter of 
exposure and mountain shadow. Because cloudiness does not 
admit of generalization one must always make a local study. 
Besides the general considerations of latitude, continentality, 
and winds, one must arrive at the local point of view through 
study of exposure, local relief, slope, and mass of mountain. 



Local Winds 

The question of local winds, like the question of local clouds, 
so varies according to locality that generalizations are out of the 
question. The best that can be done is to describe some actual 
occurrences of typical winds. 

On the windward side of a mountain barrier valleys may have 
winds of a direction reverse to the prevailing wind of the mo- 



utumn 


Year 


6.2 


6.1 


5-2 


5-2 


4.2 


4.4 


5-5 


5-4 


6.2 


6.5 



WINDS, CLOUDS, AND SUN 



6 7 



merit. The apparent explanation of a valley wind of a direction 
opposite to the prevailing wind of the unobstructed upper layers 
is that the lower edge of the wind is caught by the projecting 
mountain crest and forced down and backwards along the valley. 
This is much the same as a back eddy at the side of a river. Such 
a wind is illustrated in the valley of Bagneres-de-Bigorre in the 
central French Pyrenees. This town has an altitude of 550 me- 
ters. Seventeen kilometers up the valley is the Pic du Midi, 
whose summit is 2877 meters in elevation. Both town and peak 
have meteorological stations, and indeed the summit station is 
reputed to be the highest permanently inhabited house in 
Europe. The valley has a north-south trend. The range crest 
runs east-west and the Pic du Midi is an outlier separated from 
the crest. 

From the records a sequence of 20 days was selected at ran- 
dom. 



Contrasted Winds of Valley and Peak 



March 1910 


Pic du Midi 


Bagnfcrcs 


I 


NW 


NW 


2 


NE 


S* 


3 


NE 


s* 


4 


W 


SE* 


5 


sw 


S* 


6 


ssw 


S* 


7 


NW 


S* 


8 


ENE 


S* 


9 


SW 


SSE* 


10 


wsw 


S* 


11 


sw 


S* 


12 


sw 


SE* 


l 3 


NE 


NW* 


J 4 


NE 


S* 


15 


NE 


Calm 


16 


NW 


Calm 


n 


NW 


S* 


18 


W 


S* 


l 9 


w 


N 


20 


NE 


NNW 



From the table it will be seen that there are 1 8 days of wind in 
the valley. Of these, 1 5 (marked with asterisks) have the valley 
reverse wind. 

On clear days, with a serene and deep blue sky, there is fre- 
quently to be seen a pennant-like cloud which stretches out from 



68 MOUNTAIN GEOGRAPHY 

the leeward side of a peak. There are two theoretical explana- 
tions of the origin of these cloud banners commonly given in 
print. One is that the wind striking the face of the mountain in- 
creases locally the atmospheric pressure. On passing to the far 
side of the peak the wind enters a region of reduced pressure and 
condensation occurs. The second, and more common, explana- 
tion is that the face of the mountain chills the wind and the con- 
densed moisture is blown to windward. If the first of the two ex- 
planations is true, why should the cloud occur only at the peak? 
There is the same differentiation of pressure to be found lower 
on the mountain flanks. As to the second explanation, alpinists 
know well that the rock temperatures of mountain peaks at mid- 
day are quite high as compared with the air temperatures. 

The true explanation is that a wind will, upon passing a peak 
or a range, form an indraft of ascending air. At such times the 
valley wind will be exactly opposite in direction to the high-level 
prevailing wind of the day. This is a common occurrence. The 
ascending wind is not infrequently cloud-forming at the head of 
the valley. The same thing will happen at a peak. There will be, 
even though quite localized, an ascending current on the leeward 
side of the peak which forms clouds. The cloud upon coming 
within the influence of the prevailing wind will stream out from 
the peak as a banner. In figure 1 7 we have a situation which 
goes far towards explaining this type of cloud formation. This 
example was seen in the Tarentaise not far from the Petit Saint 
Bernard. The east wind coming from Italy not infrequently 
covers the Tarentaise and especially the Vanois with clouds. 
This is perhaps owing to the contrasted temperatures of warm 
Italy and somewhat cooler France. But the extent to which the 
indraft on the leeward sides of mountains contributes to this 
cloud formation may be guessed at from the diagram. Definite 
movements of the clouds showed the direction of the winds. The 
more steep the leeward side of the mountain the more definitely 
curved was the direction of the indraft. One is led to suspect 
that the steeper the leeward side the stronger the current of the 
indraft. 

A well recognized example of this indraft is a common phe- 
nomenon at the observatory of the Pic du Midi in Bigorre. The 
usual summer wind over this peak is from the southwest. The 
west wind is commonly accompanied by a strong wind up from 
the northern valley and a cloud where this wind coming up 




■HI gj/BSn 




Fig. 15. Clouds in the Col de Tourmalet, French Pyrenees 
The clouds in the valley are a stage in the clearing of weather following a general 
storm derived from clouds of a higher level. This is sheep country. The material in 
the foreground represents road repairs. The trail leads to higher pastures. 



*/. , € 




4 m 






i^, 



Fig. 16. Clouds after a Snowfall, Hohen Tauren, Austria 
Typical of clearing weather in mountain valleys. 



WINDS, CLOUDS, AND SUN 69 

(amont *) meets the prevailing wind. Almost exactly half of the 
platform upon which the Observatoire is located is then in cloud, 
according to M. Camille Dauzere, the Director. 



^^^rSeS 




Fig. 17. Indraft Winds in the Tarentaise, French Alps 

Proof extraordinary comes from the volume First over Everest by 
P. F. M. Fellowes and others (London, 1933), the story of the air- 
plane flight. The planes flew through the famous cloud banner 
of that peak. The observers reported the cloud to be a stream, 
several miles in length, of icy particles drawn up from the slope 
immediately below. It was not blown from the windward side 
like a 'smoking' sand dune. Indeed the windward side was bare 
of snow and ice. 

The writer is familiar with winds and clouds in the region of the 
Conflent and Valespir, Pyrcnees-Orientales. A remarkable in- 
draft or contrary wind is experienced in the valleys of the Con- 
flent under the influence of the southwest high-level wind which 
occurs with great regularity in summer. The normal valley wind 
under these conditions is an indraft wind from the north. This 
wind forms not infrequently a cloud at the headwaters of the 
Cadi, a north-flowing stream, and upon the summit land on the 
crest of the Pyrenees. The observations of the writer during a 
period in June, 1928, were typical of those weather conditions. 



Day 
I I 
12 
13 
14 
*5 



Example of an Indraft Wind 

Prevailing high-level wind Wind at Vernet-lcs-Bains 

SW NE 

SW NE 

SW NE 

SW NE 

SW NE 



1 The reader will do well to add to his geographical vocabulary the French words 
amont, meaning upstream, and aval, meaning downstream. 



70 MOUNTAIN GEOGRAPHY 

Under such conditions a cloud banner forms about the peak of 
Mount Canigou. The situation is much the same as at the Pic du 
Midi. Mount Canigou is a sharp peak with summit surface 
barely large enough for a refuge. On the east side is a steep 
cirque wall. Under the influence of a southwest wind there is an 
ascending indraft of considerable velocity. Standing on the peak 
at such a time one may be in warm sunlight and yet stretch one 
hand over the cirque and into cloud. This fog that rises from the 
cirque makes its way upward for some twenty meters, and then is 
sharply turned and carried off in a northeasterly direction as a 
distinct cloud banner. 

The importance of these winds and resulting clouds may be 
great. The Valespir is a valley region which runs in the direction 
of the prevailing (southwest) wind of summer. It lies just south 
of Mount Canigou and runs from the Pla de Guilhem towards 
the Mediterranean Sea. The head of this valley and the Pla de 
Guilhem are famous for the amount of summer cloud which 
they experience. The tree line on the west, north, and east of 
Mount Canigou is consistently at 2200 meters, but on the south, 
or Valespir slope, is only 1 600 to 1 800 meters. The zone of the 
mountain pine is here missing. The only explanation for the 
lower tree line on this southern slope which seemed plausible to 
the writer was that cloudiness during the growing season was so 
great that it reduced temperatures and therefore the height limit 
of the tree. Ordinary cloudiness would be easily explained, as in 
the other cases, by the indraft condensing upon ascension at the 
valley head. The extraordinary cloudiness, especially in sum- 
mer afternoons, called for another explanation. So heavy is this 
fog that foot paths across the Pla de Guilhem between the Con- 
flent and the Valespir are marked with cairns only a few meters 
apart, and even then one is in danger of losing the way. A pos- 
sible explanation lies in considering this indraft of air a continua- 
tion of an afternoon sea breeze from off the Mediterranean. 
Moist air from the sea is brought across the narrow plain of Rous- 
sillon and sucked up by the indraft or contrary wind, which, on 
coming in contact with the cool prevailing wind and experienc- 
ing adiabatic cooling, undergoes a dense condensation. 

The Mistral and Bora are forms of fall wind. The Mistral is a 
cold wind which descends with force and lowering of temperature 
from the French Massif Central to the Mediterranean plain. 
The Bora is a similar wind coming from the Dalmatian highlands 



WINDS, CLOUDS, AND SUN 71 

to the Adriatic, to the consternation of sailors. Both words indi- 
cate 'north,' but they are used regardless of direction for winds 
arising from similar topographic and weather conditions in other 
parts of the world. They are, it is usually thought, the result of 
barometric low pressure over the lowlands while the uplands are 
receiving cold air from an anticyclone. The normal adjustments 
between anticyclone and cyclone are impossible because of the 
plateau or upland area. The elevation of the upland intensifies 
the low temperatures. Though Mistral and Bora are heated by 
descent, nevertheless they arrive on the plains critically cold 
because of the low initial temperature. These winds result from 
an accumulated lack of equilibrium. A warm spell in the winter 
on the French Mediterranean littoral, if of several days' duration, 
might well intensify the Mistral when once a convectional over- 
turn began and the cold, heavy air of the massif continued its 
descent. 

The Mistral and the Bora are much feared. The lowering of 
temperature to people used to mild winters is most unpleasant, 
especially as the wind has great force. Stone walls and other 
windbreaks protect the fruit. It is said that one village on the 
southern slope of the Massif Central has been abandoned because 
of the force of the Mistral. The railway crossing the barren crest 
of the Dalmatian Alps and descending to Susak (Fiume) is 
particularly exposed to the Bora. On the north side of curves of 
the track are heavy masonry walls some five meters high as a 
protection against this wind. Generally speaking, Mistral or 
Bora winds are of greater significance to peoples on plains or seas 
than to the mountaineers themselves. 

Mountain and valley winds are phenomena of clear weather. 
The mountain wind blows down the valley between sunset and 
sunrise. The valley wind is a daytime phenomenon and blows 
up the valley, usually with much less velocity than the down- 
valley or mountain wind. The mountain wind may be zephyr- 
like, or it may attain the velocity of a gale. The wind becomes 
an c aerial torrent ' only when the valley is more or less free from 
forest, the valley floor is long and steep, and the upland and 
drainage area is relatively great. The rate of the descent of air 
and consequent introduction of low temperatures into the valley 
would be greater if it were not for adiabatic heating. 

The mechanics of the origin of the mountain wind are these. 
On clear evenings the radiation of the heat from the rocks at high 



72 MOUNTAIN GEOGRAPHY 

levels is more rapid than in the valleys, and the resultant cooling 
of air in contact with the rocks creates a relatively heavy stratum 
of air on the walls of the upper valley which then slips down the 
valley. In planning a camp in a canyon, the bedding roll should 
be spread on the up-canyon side of the fire to avoid the smoke. 

The valley wind is somewhat less simple. The heating of the 
valley bottom and walls results in a decided increase in air tem- 
perature. There is expansion of air in the valley bottom and a 
blowing-up of the isobaric surface over the middle of the valley. 
Air slides down the curved isobaric surface and forces its way up 
the valley sides. For obvious reasons a sunny slope will have a 
stronger amount of breeze than a shady slope. There is no essen- 
tial difference between a daytime valley wind and a convection 
current over a plain, except for the guiding influence of the valley 
wall. 

Valley and mountain breezes are characteristic of valleys trans- 
verse to a range. They are ordinarily not phenomena of large 
corridor valleys. Mountain winds may exist in daytime because 
of glaciers or snow-covered surfaces. The same factors would in- 
crease the intensity of the night-time mountain wind. 

So persistent and forceful are mountain winds in the valleys of 
the Tarentaise that the trees lean leeward. In Tibet and Kashgar 
these winds are said to make travel most difficult. On the other 
hand, they give pleasant ventilation to valleys on hot summer 
days, and they serve to lessen the contrast of day temperature 
between sunny and shady slopes. 

The Foehn Wind 

A Foehn 1 wind is a movement of air across the mountain dur- 
ing the passage of a cyclonic low on the lee side of the range. It 
has a strong development in the Alps during the passage of cy- 
clones over the North Sea area. 

The typical Foehn wind then approaches the Alps from the 
south. In ascending the mountain it gives out rain. The latent 
heat of condensation warms the air and counteracts in part the 
adiabatic loss of heat occasioned by the ascent. The wind arrives 
at the mountain crest without the complete loss of heat which is a 
measure of expansion at the summit elevation. There is no rea- 
son why the heat of descent should not be completely taken on by 

1 The American term is Chinook. 






• - • •• 




■ -v^i " .'- .- i 



Fig. i 8. Emerging from the Clouds, Spanish Sierra Nevada 
On this day the peaks resting in sunlight were invisible from valley stations. 




Fig. 19. Valley Clouds in Winter, Karawanken Alps, Austria 



WINDS, CLOUDS, AND SUN 



73 



the wind as it descends into the lee valley. A Foehn wind there- 
fore arrives hot and dry on the north slope stations (Figures 21 
and 22). * 

Hann was among the first to state that the Foehn was not, as 
popular belief would have it, the hot breath of the Sahara, but 
appeared on the Italian slope as a rain wind. Ficker and Bill- 
wilier are the most searching students of the Foehn today. 

The reality of the Foehn is easily discoverable by coincident 
observations on two sides of the Alps. The following table com- 
pares an Italian station at the same time as an Austrian station: 

Foehn Weathers on Opposite Slopes of the Alps 

Temperature Relative humidity 

Milan 3.2 C. 96% 



Bludenz 



.i°C. 



29% 



At Bludenz on February 1, 1869, trie temperature was found to 
be 19.3 degrees and the relative humidity 14 per cent. This 
temperature was 15.7 degrees above normal for that day, whereas 
the relative humidity was 58 per cent below normal. The suc- 
cession of weathers during a Foehn at a single station is shown by 
the following data: 



Wlather During a Foehn at Zurich 











Nov. 28, 


1897 










I A M 


3 


5 


7 


1 1 


I P.M. 


3 


5 


Press. 


72I.O 


20.7 


19-9 


19.4 


18.1 


l6.2 


14.6 


13-3 


Temp. 


O.4 


1.6 


°-3 


-0.5 


4.8 


8-3 


I O.I 


9-3 


R. H. 


75 


62 


76 


74 


44 


29 


20 


16 


Wind 


N 


NW 


VVNW 


NW 
Nov. 29, 


WNW 
1897 


WNW 


SW 


wsv 




7 


9 


1 1 


I A M. 


3 


5 


7 


9 


Press. 


1 1 -3 


9-3 


8.0 


6.0 


4-3 


i-9 


99-9 


2.8 


Temp. 


8.4 


9-6 


10. 1 


6. 9 


5-i 


6.7 


6-3 


2.4 


R. H. 


12 


5 


18 


62 


84 


78 


82 


85 


Wind 


wsw 


WSW 


w 


WSW 


WSW 


SW 


W 


w 



The importance of the influence of the Foehn depends greatly 
upon the details of exposure. One field will have vines while the 

1 "If the Foehn does not blow 

The good God and the warm sun 

Can do little with the snow." 
"The Foehn can achieve more in two days than the sun in ten." 
"Tonight the wolf is going to eat the snow." 



74 



MOUNTAIN GEOGRAPHY 



next, deprived of the influence of the Foehn, will not have suffi- 
cient warmth for vines. This localization is demonstrated by the 
higher temperatures at Altdorf, which, though lower in elevation 
than Zurich, is better exposed to the Foehn. The difference in 
elevation is not sufficient to account for the difference in tempera- 
ture. 

Importance of Exposure to Foehn 

Altitude Winter Spring Summer 

Zurich 470 m. -0.3 8.9 17.6 

Altdorf 454 1.1 9.5 17.3 

Of importance equal to the raising of temperatures is the dry- 
ing effect of the wind. Under the Foehn influence the snow is 
quickly evaporated and the ground dried and 'sweetened' for 
early spring plant growth. The vine and maize are grown in cer- 
tain localities of Grisons and Vorarlberg. Their culture is con- 
fined almost exclusively to the areas exposed to the Foehn. The 



Fall 


Year 


8.8 


8.7 


10.8 


9-5 




Fig. 20. Facts of the Foehn Wind 
(Courtesy of Ginn and Company, New College Geography.) 

same cause in part explains the grape culture which produces the 
fondant de Ston, the best Swiss wine. 

The writer's personal experiences with Foehn winds have been 
numerous. Once on a train trip from Lausanne to Milan via the 
Simplon a rapid cross section of the Foehn was made. Going up 
the Rhone Valley the weather became hot and dry. The low 
relative humidity of the air was shown by the clarity of long dis- 
tance vision. After leaving Sion in heat and dust, one discovered 



WINDS, CLOUDS, AND SUN 



75 




T, T, T T s 

Fig. 21. Mechanics of the Foehn Wind 

The altitude (H) equals elevation and the base (T) the temperature. As air rises 
to the level H adiabatic cooling decreases the temperature from T to T^. At H 
precipitation takes place and the rate of cooling is reduced because of liberated heat 
of condensation (6°C per kilogram of water). This rate is the line AH. On descend- 
ing the air has adiabatic heating at the same rate as indicated by TA. The air has 
then on arriving at the lee base of the mountain the Foehn temperature of T 3 . 



76 MOUNTAIN GEOGRAPHY 

on emerging from the Simplon the Italian countryside in a deluge 
of downpour ing rain. The writer underwent another typical 
experience with the Foehn wind, best told in the first person: 

During an evening in an Innsbruck Beerstube with an Austrian friend, 
I complained of a sense of nervousness. The people about me likewise 
were temperamental. The Austrian explained that it was the Foehn a 
thousand meters above, and that it would be down in the valley in the 
morning. His surest way to forecast Foehns was by the temperamental 
expressions of people. "A political Putsch" he said, "in Tirol is al- 
most invariably at the time of Foehn." The next day I saw in a book 
store the title Menschen in Fohn, and I guessed its contents. 

I was advised to go the next day into mountain villages in the Stu- 
baiertal to see the action of the Foehn. The day was remarkably clear. 
Trees on the distant mountains stood out with complete definition. 
The crest of the Alps over the Brenner Pass showed a great cloud, 
the Foehn bank, indicating the edge of the Italian rain storm. The 
streets of the towns were deserted by man and beast. The shutters 
flapped in the wind. People stayed indoors. Indeed, so dry and gusty 
is the wind that certain towns have regulations against the building 
of fires at the time of Foehn. My wife and I returned to Innsbruck 
utterly exhausted by the weather. 

The Chinook wind is the American Foehn. It is usually a west 
wind. The name came from its early description as occurring in 
the region of the Chinook Indians of Washington. Its most 
significant appearance, however, is when it blows down upon the 
Great Plains along the margins of the Rocky Mountains. It 
occurs in the Rockies from Colorado to Alberta. It makes its ap- 
pearance suddenly and may raise the temperature many degrees 
in a few minutes. Ballou reports a rise from — 10 to 20 degrees C. 
in a day. It is known as the snow-eater. Cattle may be in stable 
one day while pastures lie under snow. In twenty-four hours 
these cattle may be eating grass out of doors. It may blow con- 
tinuously for many days, during which special fire laws prevail 
because of the low relative humidity. 

BIBLIOGRAPHICAL NOTES 

Wind, Cloud, and Sun 

One should consult the Note on General References on Mountain Climate, 
especially Hann, Humphreys, and Benevent. Benevent is particularly 
clear. Hann is the great classic, and has an exhaustive bibliography of 



WINDS, CLOUDS, AND SUN 77 

German references. Humphreys is theoretical. Most of the regional 

studies will be of value here. 

Ballou, H. M. " The Chinook Wind," in American Meteorological Journal, 

ix (1892-93), pp. 541-547. With a bibliography. 
Billwiller, Robert. " Uber verschiedene Entstehungsarten und Er- 

scheingungsformen des Fohns," in Meteorologische Zeitschrift, xvi 

(1899), pp. 204-215. 
Blair, W. R., and Ross, L. C. " Stationary Clouds to the Leeward of 

Hill and Mountain Ranges," in Bulletin of the Mount Weather Ob- 
servatory, ii (1909-10), PP- 75-77- 
Burrows, A. T. " The Chinook Winds," in United States Department of 

Agriculture, Yearbook, 1901, pp. 555-566. 
Davis, W. M. "The Foehn in the Andes," in American Meteorological 

Journal, iii (1886-87), pp. 507-516. 
Defant, Albert. "Berg- und Talwinde in Sudtirol," in Sitzungsberichte of 

the Vienna Academy, math.- nat. Klasse, cxviii, 2 (1909), pp. 553- 

604. 
Defant, Albert. "Das Klima von Innsbruck mit besonderer Rucksicht 

auf den Fohn," in Deutsche Rundschau fur Geographie, xxxiv (191 2), 

PP- 405-410- 
Dersch, Otto. " Uber den Ursprung des Mistral," in Zjeitschrijt der 

Osterreichischen Gesellschaft fur Meteorologie, xvi (1881), pp. 52-57. 
Douglas, C. K. M. "Some Alpine Cloud Forms," in Quarterly Journal 

of the Royal Meteorological Society, liv (1928), pp. 175-178. 
Ficker, H. von. Fickcr has written extensively on the Foehn, and is an 

authority. One should see Denkschriften of the Vienna Academy, 

math. -nat. Klasse, lxxviii (1906) and lxxxv (1910); Meteorologische 

^eitschnft, 1905, 191 o, and 191 1; and ^eitschrift des Deutschen und 

Osterreichischen Alpenvereins, xliii (191 2), pp. 53-77, "Die Erforschung 

der Fohnerscheinungcn in den Alpen." 
Hales, W. B. "Canyon Winds of the Wasatch Mountains," in Bulletin 

of the American Meteorological Society, xiv (1933), pp. 194—196. 
Hann, Julius. " Bewolkung und Sonnenschcin auf dem Sonnwendstein 

(1470 m.) bei Wien," in Meteorologische ^eitschrift, xxxiii (191 6), pp. 

554-556. 
Hann, Julius. " Uber den Fohn in Bludenz," in Sitzungsberichte of the 

Vienna Academy, math. -nat. Klasse, lxxxv, 2 (1882), pp. 416-440. 
Hann, Julius. "Zur Theorie der Berg- und Thalwinde," in ^eitschrift 

der Osterreichischen Gesellschaft fur Meteorologie, xiv (1879), pp. 444- 

448. Classical. 
Huber, Anton. "Der Gang des Sonnscheins auf der Zugspitze und in 

Partenkirchen," in Deutsche s Meteor ologischen Jahrbuch fiir Bay em, 

1930, pp. G i-G 13. 
Huber, Anton. " Sonnenschein an Niederschlagstagen," in Beitrdge zur 

Geophysik, xxxiii (1931), pp. 1 18-124. 



78 MOUNTAIN GEOGRAPHY 

Loud, F. A. " The Diurnal Variation of Wind-Direction at Colorado 

Springs," in American Meteorological Journal, i (1884-85), pp. 347- 

354. 
McCaul, C. G. " South Alberta and the Climatic Effects of the Chinook 

Wind," in American Meteorological Journal , v (1888-89), pp. I45~i59> 

362-369. 
Martonne, Emmanuel de. "Note preliminaire sur le vent d'autan," in 

Bulletin de la Societe Languedocienne de geographic, xxx (1907), pp. 100— 

114. 
Peattie, Roderick. "Nuages en banniere: Petite 6tude des vents et des 

nuages de montagne," in Revue de geographie alpine , xvii (1929), pp. 

329-335- 
Peppier, W. " Zum Einfluss des Fohns auf die Mitteltemperatur im 
Alpenvorland," in Meteor ologische Zjeitschrift, xliii (1926), pp. 374- 

375- 

Pernter, J. M. Uber, die Haufigkeit, die Dauer, und die meteoro- 
logischen Eigenschaften des Fohns in Innsbruck," in Sitzungs- 
berichte of the Vienna Academy, math.-nat. Klasse, civ, 2a (1895), 
pp. 427-461. 

Stupart, R. F. "The Chinook in Southern Alberta and Temperature 
Inversions at Sulphur Mountain, Banff," in Proceedings and Transac- 
tions of the Royal Society of Canada, 3d ser., iv (1910), section 3, pp. 
51-52. 



CHAPTER IV 

VEGETATION ZONES AND THE HEIGHT LIMITS 

OF FIELDS 

Mountain Zones 

ONE of the most attractive concepts in the study of mountains 
is the conception of zones. Travelers among mountains de- 
light in telling how their climb began amidst a splendor of tropi- 
cal foliage. Then in succession they went from the evergreen 
broad-leafed zone to that of the deciduous trees, to the evergreen 
conifers, to an arctic heath, and so to eternal snow. The rapid 
contrasts of vegetation within so small a space challenge the 
imagination. (Figure 22.) 

There are many ways of describing and defining zones. The 
most common type is the following: l 

Vegetation Zones in Northeastern Colorado 

1. Plains (grass-steppe or short grassland), up to 1830 meters. 

2. Chaparral or brushland of chokecherry, thornapple, mountain 
mahogany, etc. A narrow, interrupted belt. 

3. Yellow pine, Douglas fir zone, 1830-2440 meters. 

4. Lodgepole pine zone, 2440-3050 meters. 

5. Engelmann spruce, balsam fir zone, 3050-3500 meters. 

6. Alpine zone. Above timber line. 

Season Zones in Alps 
In meters 

500- 650- 1000- 1300- 

600 1000 1300 1600 

Awakening of vegetation ... Mar. 17 Mar. 30 Apr. 10 Apr. 21 

Cherries blossom May 5 May 10 May 16 May 21 

Hay harvest June 15-20 June 24 June 25 June 27 

Cherries ripen June 25 July 18 Aug. 3 Aug. 20 

Winter wheat ripe July 18 July 31 Aug. 8 Aug. 18 

Oats ripen Aug. 14 Aug. 27 Sep. 5 Sep. 16 

Ceneral snow cover Dec. 10 Nov. 30 Nov. 20 Nov. 10 

1 Robbins. 



8o 



WINDS, CLOUDS, AND SUN 




A A " A A A 

A* a A 

A * * * A 

JV A 



A 

A a A A 
A A 

^\> 

FOB6E '» /» 




12 3 4 

KILO MtT£R5 



Symbols 


zr~z: 


: -v:| 


rt£Los 


PASTUei 


W 


A W 



foblst pasture, 

in forest 



Fig. 22. Simple Zoning in the Valley of Llosa, Spanish Pyrenees 

{After Sorre.) 



HEIGHT LIMITS OF FIELDS 



8l 



Season Zones in Alps (continued) 
In meters 



1600- 
2000 

May 2 
June 21 
July 1 

Sep. 3\ 
Sep. 29/ 
Oct. 28 



2000- 
2300 

June 2 
July 11 
Aug. 3 



[690 meters 
Oct. 15 



2300- 
2600 

June 28 
July 29 



Oct. 



Here retardation in blossom time is 10 days for 300 meters and 
the retardation in fruit time is 12.5 for the same difference in 
elevation. Of the same type is the following data from Andorra. 
The chestnut and walnut are found only in the neighborhood of 
the lowest village, Sant Julia. The live oak occupies sunny 
slopes in the basin of Andorra la Vieja. The climate, always 
rigorous, increases in severity very quickly with altitude. Many 
valleys are entirely subalpine and alpine in character. In less 
than three hours of valley way one may pass from harvest rye to 
green rye. On the plain of Andorra (1000 meters) the grain is 
harvested from July first to the tenth. At Encamp (1600 meters) 
the harvest is nine days later. At Lo Serrat (1600 meters) harvest 
is not until the twentieth of August, when the peasants have al- 
ready begun plowing some fields for the next season's crop. 

Duration of the Dormant Period of Vegetation in Andorra 











Duration 


Vegetation 

Broadleaf 


Locality 

Plain of Andorra 


Beginning 


End 


(months) 


evergreens 


, (1029 m.) 


Dec. 15 


Jan. 15 


I 




fLa Massana 


Dec. 1 


Feb. 15 


*t 




(1252 m.) 








Deciduous 


Encamp (1266 m.) 


Dec. 1 1 


Feb. 15 


2-t 




^Ordino (1304 m.) 


Dec. 15 


Feb. 30 


*i 




Soldeu (1825 m.) 


Oct. 15 


Apr. 15 


6 




Cortals d'Encamp 








Coniferus 


(1860 m.) 


Oct. 15 


Apr. 15 


6 


subalpine 


Lo Serrat (1600 m.) 
Bordes d'Incles 


Oct. 1 


Apr. 30 


7 




, (1825 m.) 


Oct. 1 


Apr. 30 


7 




Port de Soldeu 








• 


(2407 m.) 


Sept. 25 


June 1 


si 


Alpine 


Portclla Blanca 










(2515 m.) 


Sept. 25 


June 15 


8f 



82 MOUNTAIN GEOGRAPHY 

Zones at best illustrate an average condition. There are, 
nevertheless, in the broader sense, belts of vegetation along moun- 
tain sides, and there are, likewise, belts or zones of economic ad- 
justment of mountain slopes to man's use. These we shall call 
culture zones. An example of culture zones among tropical 
mountains is given by Bowman: 

Height Limits of Zones in Peru 

Snow 1 7,000 feet 5^5 meters 

Potatoes 14,000 " 4270 " 

Barley 1 3,000 " 3965 

Wheat 1 2,000 " 3660 " 

Corn 10,000 u 3050 " 

Sugar 8000 " 2440 " 

Banana 6000 " 1830 " 

This type of data demonstrates climatic limits but not climatic 
optima. Dainelli shows, in a study of an isolated Italian peak, 
that though the chestnut is found at 1 250 meters, it is important 
only between 700 and 1000 meters. Alas, however, our data on 
culture limits are ordinarily on extremes rather than optima. 

Fritsch in his work on the Ortler Alps gives the altitudes of the 
economic zones, indicating the exposure upon which the observa- 
tion was made: 

Limit Elevations in the Ortler Alps in Meters 

Maximum Minimum Mean 

Permanent habitations 1664 SW 1 147 N 1 377 m. 

Grain limits 1642 u 1207 NW 1390 

Mown fields 2 1 08 " 1 474 N 1 767 

Herdsmen's cottages 2 154 " 1 757 NE 1952 

Forest line 2 1 59 " 2023 NE 2 1 1 8 

Shepherds' huts 2342 " 2065 W 2189 

Tree line 2323 NW 2166 NE 2253 

Orographic snow line 2754 S 2533 N 2629 

Climatic snow line 3089 S 2854 N 2964 

Let us consider a series of zones on the flanks of Mount Cani- 
gou. Mount Canigou is a relatively isolated peak in the eastern 
French Pyrenees. Its peak, 2785 meters in elevation, is within 
sight of the Mediterranean. Sorre would have three climatic 
levels on the Mediterranean flank, each with characteristic vege- 
tation. Up to 700 meters vegetation is conditioned by the need 
of conserving a limited supply of water. Winter cereals, the olive, 



HEIGHT LIMITS OF FIELDS 83 

and the evergreen oak flourish. From 700 to 1 700 meters there is 
an abundance of well distributed rain and an increased rigor of 
winter. Here are hay fields, rye, conifers, and the beech tree. 
Above is a forest and pasture zone, inhabited only in summer, 
where there are lower temperatures, greater precipitation, and 
considerable snow cover. But only in a general sense does one 
find zones on this peak. 



bO'F 



50 

















•^u« 




4/^ 


















30 


oo' nc 


00 S0O0 bOOO 71000' 



Fig. 23. Reversal of Vegetation Zones 

The reversal of vegetation zones through interlocking of spur and valley. Thus if 
50.6 degrees F. is a critical temperature for a plant, such temperature is found at 
7000 feet on the spurs. By following the abscissal direction the same condition is 
discovered in the valley at 4650 feet elevation. (AJler Shreve.) 

Shreve, in his excellent study of the Santa Catalina Mountains 
of Arizona, gives a description of zones well worth retelling. A 
journey from the base to the summit of these mountains im- 
presses upon one the constant changing of plant types. There is 
no portion of the mountain slopes, at least below 7500 feet, where 
a climb of 500 feet does not materially alter the physiognomy of 
the vegetation. The cacti and desert shrubs give way to ever- 
green oaks, leaf succulents, schlerophyllons, shrubs, and plants. 
The upper limits of the desert are 4000 feet on the north slope 
and 4500 feet on the south. The lowest limit of the true forest is, 
on the north slope, at about 5800 feet, while on the south slope it 
is about 6300 feet. The mountains here are not high enough to 
attain the climatic forest line. No native plant is able to adjust 
itself to the range of conditions found within the 6000 feet. Most 
plants here do not have altitude ranges of more than 1000 feet. 



84 MOUNTAIN GEOGRAPHY 

The climatic factors involved in the determination of a normal 
altitudinal gradient of vegetation are of three groups. They are 
moisture factors, temperature factors, and light factors. The 
moisture factors include not only zones of rainfall but duration 
and thickness of snow cover. Negatively they include evapora- 
tion, which is, of course, a function of wind, temperature, soil, 
and vegetation. The temperature factors, like the previous 
group, imply conditions of relief, as exposure and angle of slope. 
Temperature has a more direct relation to altitude than mois- 
ture, but even then generalizations are dangerous. Light factors 
enter into plant characters. Temperature at high altitudes is 
dependent largely upon direct insolation. The soil factor is for 
the moment left out of consideration. 

With three sets of variables, all of them interrelated, a variety 
of conditions will arise. Moreover, the quality of the meteoro- 
logical element as well as its numerical total or mean is impor- 
tant. Mount Carmel, on the Palestinian coast, has no great 
rainfall, but because of the great amounts of dew and fog and 
reduced evaporation it is green throughout the year. This is, of 
course, in contrast to the brown lowland of Palestine. 

The full appreciation of the variables weakens one's faith in 
the idea of mountain zones. Krebs in his great study of the 
Eastern Alps finds it necessary to abandon the use of zones. 
Blache in a detailed study on the Pre- Alps of Savoy is cautious in 
speaking of zones because of their lack of sharp boundaries. He 
refers to them only in a general manner. Shreve says, "The 
occurrence in nature of irregularities of relief is responsible, how- 
ever, for local departures from the ideal vertical gradients of 
climate and also from the ideal altitudinal distribution of vege- 
tation which would be anticipated on a geometrically constructed 
mountain." 

The value of the conception of zones in a study of mountain 
geography lies in the perspective we may obtain in the general 
relations between mountains and modifications of climate. Thus, 
for example, the importance of latitude as a factor in climate is 
brought out with a clearness which could not be obtained were 
we to break down the broader idea of zones by laying most stress 
upon details which mark exceptions. Zones on mountains lead 
us to such generalizations as may exist on the relationship be- 
tween altitude and temperature. On the other hand, the further 
we proceed in details of the study of mountain vegetations and 



HEIGHT LIMITS OF FIELDS 85 

altitudinal arrangements of economies the less important does the 
zone conception become. 

Height Limits of Fields 

The determination of the factors controlling height limits of 
cultures, particularly agriculture, is one of prime importance in 
the economic geography of mountains. It is difficult to gen- 
eralize in the matter of elevation of field limits. Each field in 
mountainous terrain has its distinctive climate, soil, degree of 
accessibility, and economic history. Numerous field observa- 
tions are here unrecorded because they have each a particular 
rather than a general value. We shall do well, however, to discuss 
what is meant by a height limit. Are we to take an extreme figure 
or an average figure? Ellsworth Huntington would probably 
favor the average height limit figure for fields or for a single crop 
as approximating an optimum. The discrepancy between the 
maximum field height and the average field height is ordinarily 
not greater than 200 meters. Otto Lehmann in his article on the 
heights of dwellings has two agricultural limits \jarm limits •, that 
is the limit of continuously enclosed agricultural regions, and 
field limits^ which include tilled islands beyond the area of con- 
tinuously enclosed agricultural regions. 

Factors of slope, accessibility, distance from sources of manure, 
density of population, and local agricultural habits and eco- 
nomics also influence height limits of fields. Ratzel calls this 
altitudinal frontier of tillage the Kampsaum. The actual limit is 
determined by observation, but the factors are partly economic 
and partly climatic. More often the type of agriculture has a 
climatic limit rather than the agriculture itself. Thus in the 
Stubaier Alps the grains are confined to the valleys and lower 
slopes and the determining factors are climatic, yet potatoes and 
turnips are grown in fields far above the normal agricultural 
zone. Barley for grain is limited about Zermatt, Switzerland, by 
the 2100 meter contour. Barley for straw for cattle bedding is 
produced above that contour. 

Though we are unable to discuss the generalized data of height 
limits of fields because of the variety of factors, we shall be able 
to analyze the character of the factors. Altitude per se has no im- 
portant effects upon the cultivated plants, and though the cli- 
matic factors indirectly are of prime consequence we shall be able 



86 MOUNTAIN GEOGRAPHY 

to ignore the direct importance of diminution of pressure. This 
chapter will consider in turn topography, exposure to the sun, 
and geology as factors in the height limits of fields. The special 
study on Andorra is of both topographical and climatological 
significance. The study of the Conflent stresses accessibility. 
The study of the Doron brings out topographic economic fac- 
tors. The importance of the economic factor is tersely stated by 
Blache when he says that if an inhabitant of a particular valley 
wishes to content himself with a mediocre harvest he will culti- 
vate the vine, wheat, maize, and fruit trees up to a considerable 
altitude. If, however, the commercial element becomes pre- 
dominant, the marginal zone will be abandoned and the limit of 
cultivation will recede. 

Topography affects cultivation limits in several manners. 
First, in the alternate distribution of valley and spur it creates an 
interlocking of contrasted climates. Secondly, the contrast of 
level land and slope so steep as to be prohibitive for field culti- 
vation is a matter of great significance; but it is too obvious to 
need elaboration except in its climatic implications. 

One of the chief irregularities in the general zonal conception 
of mountain plant distribution is due to this interlocking of spur 
and valley (Figure 23). Because of this there may occur an in- 
version of order of plant associations. Plants of a lower zone 
may be found at a higher elevation on the floor of a well warmed 
valley than plants of the upper zone on the exposed mountain 
spur. An observation on a plant elevation should be made always 
in the light of the topographic situation, the angle of slope, ex- 
posure, nature of slope, total elevation, and mass of mountain. 
Mayer in his excellent article on the physiographic basis of 
height limits says that field limits on slopes are usually expressive 
of topographic conditions. He would have us find the climatic 
limitation always along the valley bottom. Indeed, the two are 
interrelated, especially when we consider water tables. Cer- 
tainly it is true that the height of fields in many mountain areas 
represents the height of valley bottoms. In this regard it must be 
remembered that the height of valley bottoms is a function of 
mass of mountain. Other considerations to be taken into ac- 
count are that the steepness of the valley walls is commonly a 
measure of the severity of glacial erosion. The U-shaped corri- 
dor valley, other things being equal, will have lower slope tillage 
than the broad, gentle-sloped valley. In the Alps the pastures 



HEIGHT LIMITS OF FIELDS 87 

ordinarily stop short of the peaks because of the steep cliffs which 
mark the upper mountain reaches. On the other hand, the 
rounded domes of Central Sicily are farmed to the top because 
the slopes at the summit are gentle. 

Not only do the main valleys alternate with the main spurs, but 
some valleys have interlocking of side valley and side spur. 
Again there may be inverted culture zones. An enclosed side 
valley may have flanking walls so close together because of a nar- 
row valley bottom that there is a definite raising of temperature 
due to heat reflection. Moreover, side valleys are hotter because 
of the mass of mountain in which they are embedded. The color 
of rock also influences the amount of reflection. 

Cultivation does not always prefer level land. Though the 
matter is discussed later, let us say here that a sunny slope may 
be more valuable than the level land because of the more direct 
angle of the sun's rays. Not only does a favorable slope receive 
greater warmth during the maturing season, but one which re- 
ceives rays at a favorable angle in the spring will lose its snow and 
become warmed earlier. Soil temperatures are favorable for a 
longer time on a sunny slope than in a valley bottom. Moreover, 
the slope has the advantage of early draining. Valley bottoms in 
glaciated valleys may remain wet during the summer so as to 
preclude all but hay culture. This is due not merely to the irregu- 
larity of glacial deposits but also to torrential fans or cones which 
commonly block the stream course in over-steepened valleys. In 
addition to this, the drainage of cool air to the valley bottom may 
create frost pockets and the slope for this reason alone will have a 
longer growing season than the valley bottom. On the other 
hand, glaciated valleys frequently have slopes so steep that cul- 
tivation or forests cannot creep up the valley walls. This is a rea- 
son for the greater height limits of cultivation in valley courses 
than on valley sides. 

The character of the crop likewise may be different from that 
on the valley floor. Grass thrives well in the moist, shady valley 
bottom. Vines are a dry-land crop, needing little or no irrigation 
and much sunlight. They grow well on slopes, and because of the 
character of cultivation fit in well on terraces. Mown crops are 
found on slopes because, unless the ground is too steep for a foot- 
hold, it is actually easier to mow with a scythe on a slope than on 
the level. Very steep slopes are devoted to grass, if the rainfall is 
of the proper type, for the roots prevent soil erosion. (Figure 48.) 



88 



MOUNTAIN GEOGRAPHY 



The Sunlight Factor 

The matter of sunlight is so important that each mountain 
speech or dialect has its special terms for the sunny and shady- 
slopes. Some of these are herewith given: 





Sun 


Shade 


German: 


Sonnenseite 


Schattenseite 




Sonnenberg 


Schattenberg 


French: 


Adret 


Ubac 




Endroit 


Envers 


Catalan: 


Sola 


Baga 




Solatia 


Ubach 




Soula 


Ubago 


Italian: 


Indretto 


Inverso 




Adritto 


Opaco 



Frequently villages straddle streams. That portion of the vil- 
lage which is on the shady bank is less desirable for living quar- 
ters. There are, for example, Envers Fontenille and Inverso 
Pinusca. A farm which is famous as a rendezvous for epicures in 
the Provengal hills is known as 'le Ferme de PUbac Foron. 5 
Grano calls the shady side of a mountain the versant de la nuit. Be- 
cause 'sunny slope' and ' shady slope 5 are tongue-twisters, the 
French terms adret and abac will here be used in their stead. 

In considering the land values of mountain fields, a concern of 
the first importance is exposure to the sun. (Figures 4, 24, and 
25.) It is the field in the sunlight that brings a high price. A 
Catalan saying distinguishes between the two slopes in the fol- 
lowing manner: " A solane is a portion of a commune conceded to 
a cuart (a section of a commune) for pasture. Anybody has a 
right to go to the ubach" Indeed, in general, mountain people 
live in the sun. In the east- west Tarentaise valley of the French 
Alps downstream from Bourg-Saint-Maurice 89.6 per cent of the 
population lives on the adret. 

The examples of exposure contrasts in geographic literature are 
legion. Let us turn to a description of the Val de Conches, which 
is a high valley in Valais. The valley has been severely glaciated 
and has slopes of thirty degrees. The contrast here between adret 
and .ubac is great. Fifty-seven per cent of the unproductive land 
and 61 per cent of the forest is found on the ubac. The adret has 70 
per cent of the grass land and 97 per cent of the total tilled land. 







!^ » j 



' v -~ 



Fig. 24. Sunny and Shady Slope in the Tarentaise, French Alps 



^m^MmM 



tfiilHIM 



-•-it. 



Hi H 



XPHG 



Hi 






sgBHWr 



JIp 



;r~ f\i---V! i^,ft.#r &? •« - -•• V ,j* ^ HsfaKS ! « fill **• mm 

HHBHHKfP t» 11111 - HP! Al^^^^HftftgwWfj'?'- ^* i 



i SHPPR 











■HM 



Fig. 25. Agricultural Limits, Grisons, Switzerland 
These highest tilled areas on the road to Arosa are scanty potato patches. 



HEIGHT LIMITS OF FIELDS 89 

T^e 3 per cent of the cultivation on the ubac is really on secondary 
adrets. Fliickiger states that in the Loetschen valley of Switzer- 
land the southerly exposure is in pastures, fields, and settlements, 
while the ubac, the northerly exposure, is almost exclusively de- 
voted to forests. Fritzsch gives us the following contrasts in the 
Ortler Alps: 

Exposure in the Ortler Alps 

Adret Ubac 

Alpine huts 2154 m. 1757 m. 

Forests 2 159 2023 

Trees 2240 2 1 66 

An unusual study of hours of sunlight in a single valley is that of 
Levainville writing on the Barcelonnette Valley of the French 
Alps (Figure 26). He shows in this study a surprising fact, 
namely, that the ubac there actually has greater duration of sun- 
light than the valley bottom. Five stations on the adret had a pos- 
sible mean daily insolation of 597 minutes. Five stations on the 
valley bottom gave an average of 449 minutes of sunlight. Yet 
an average of four stations on the ubac received sun for an aver- 
age of 545 minutes. Duration of sunlight in a valley station is, 
then, not only a matter of latitude, height of barrier, and height 
of station, but also of detail of slope, trend of valley, and detail of 
barrier crest topography. 

The economic aspects of sunlight should be thought of in terms 
of seasons. It is relation of sun to field in growing season that 
counts. Adret and ubac near Innsbruck have a growing season of 
15 days difference in length. 

The quantitative importance of an adret position rather than a 
site on a level valley bottom is indicated by the location of the 
most elevated fields in the drainage basin of the Doron, Taren- 
taise. They are found in the hamlet of Fontanette at 1680 me- 
ters. Some hundred meters below is the flat valley bottom of 
Pralognan. This land is devoted entirely to hay. At Fontanette 
there is, however, a small adret slope. Here on the lowest level of 
the alp there are potato fields (Figure 29). 

The best resume of the story of adret and ubac is that of Marcelle 
Vessereau. It is an excellent article laying stress upon sunlight as 
a factor in height limits to the exclusion of other factors which 
modify, and in some cases counteract, the effect of insolation. In 
her conclusion she says rightly, " The law of the adret and ubac is a 



go MOUNTAIN GEOGRAPHY 

general law." The exceptions, however, would seem to be so 
many that one should apply the law not generally but specifi- 
cally, and then only after taking note of the local modifications 
and exceptions. An exception to the law of exposure is found in 
the existence of sunny slopes on an ubac. These slopes are the 
'secondary adrets.'' 



MORE THAN 9—10 

-■-— ■- rS 



10 HOURS 




10 KILOMETERS 



4* EAST OF PARIS 



Fig. 26. Sunlight Hours in the Valley of Barcelonnette, French Alps 

The irregularity is due to secondary adrets. (After Levainville.) (Courtesy of the 
Geographical Review, published by the American Geographical Society of New York.) 



Is there a difference in the insolation values on an east and 
west slope? There is, but the difference is not great. The writer 
made a study in the valley of Allues, drained by a suspended 
stream which runs north to join the Doron at Brides-les-Bains in 
the French Alps. Its two valley walls are of the same declivity 
and the crest lines of the flanking mountains are approximately 
of the same altitude. Here may be studied the relative effects of 
morning and afternoon sunlight. Cultivation on the west-facing 



HEIGHT LIMITS OF FIELDS 91 

or afternoon slope is higher by one hundred meters than on the 
east-facing or morning slope. This is normal, as the morning sun 
is frequently weakened by haze. But the local inhabitant says 
that the discrepancy in culture lines to be noted at the head of the 
valley is distinctly a matter of inferior soil on the morning slope. 
Here is another case where one might easily draw hasty and 
erroneous conclusions. 

Blumer gives us £ study of a north-south range with all the 
contrasts of wind, precipitation, and evaporation that might be 
expected. But the range, according to that author, has an in- 
solation difference for the two slopes. The east represents the 
ubac and the west the adret. The range considered is the Rincon 
Range of Arizona (2100 to 2400 meters). After describing the 
difference in vegetation on the two exposures he noted the fol- 
lowing facts. Both sides have the same type of bed rock. The 
east is the steeper slope, but it has a better humus, and the geo- 
logic structure provides better ground water conditions. But, 
also, the east slope has a sunrise three hours before the west, and 
at a time when, in this desert climate, there is heavy dew. More- 
over, the sun sets three hours earlier on the east than on the west 
slope. It sets shortly after the time of maximum temperature. 
On the west as the sun rises there is little dew on the ground. 
The afternoon sun shines on this slope during the hottest hours 
of the day. Insolation is greater on the west. Minimum tempera- 
tures are lower on the west. The summer diurnal range on the 
east approximates 8.3 to 8.8 degrees while on the west it may be 
as high as 18.3 degrees. The west slope has perhaps twice the 
summer precipitation of the east slope. The important winds are 
from the west with their attendant evaporation. On the summit 
the tips of the pine leaves all point east. 

Normally we speak only of an adret or ubac of a valley. The 
terms also may well be applied to slopes of an isolated mountain 
range. The adret of a valley and of an isolated mountain range 
are, however, not completely analogous. In a valley the differ- 
ence between the two slopes is lessened by reflected heat and by 
mixing of air by mountain and valley winds. Also the exposed 
mountain range more than the valley is open to the prevailing 
wind with its consequent precipitation or evaporation. The 
Sierra Nevada Mountains of Spain offer excellent opportunity 
for determination of the importance of insolation. Unfortu- 
nately, the writer saw them during stormy weather only. Many 



92 MOUNTAIN GEOGRAPHY 

of his conclusions are based upon data presented by Boissier, who 
made a study nearly a century ago. But the relation of adret and 
ubac is probably the same as today. 

The Sierra Nevada is an east-west range of schists and granitic 
rocks. The core is surrounded by loosely consolidated sedimen- 
taries. The geology and topography of the two slopes are identi- 
cal. Therefore these two factors can more or less be eliminated 
from our consideration. The southern slope is climatically open 
to African influences. The northern slope receives the brunt of 
the cyclonic storms and is exposed to cold winds from the Meseta 
Central. The north slope is the more moist. At the south pied- 
mont there is the orange, and, not far distant, sugar cane is 
grown. The north piedmont is the famous Vega and does not 
support the orange. The vine, walnut, olive, fig, potato, and 
maize are all found at greater heights on the south slope than on 
the north, higher on the adret than on the ubac. Boissier would 
consider greater wind and its attendant evaporation the cause 
for the lower limits on the north. Few today would agree with 
him. Most would consider the matter one of insolation. Par- 
ticularly this is true since on the north slope there is plentiful 
snow water for ground water and irrigation. 

We have suggested that soil and topography may be eliminated 
from consideration. Here in the Sierra Nevada there is no ques- 
tion of balance of land utilization. No land is reserved for do- 
mestic animals at the expense of fields. Other than a few sheep 
in the cirques there is only the household goat. The matter of 
social characteristics of the farmers on the two sides of the range 
is here not of concern, for they are the same on both slopes. 
Manure is generally lacking and therefore does not influence 
the height of fields. Accessibility, a matter so important in the 
Pyrenees and Alps, does not control the height of fields. There 
are potato fields near the snow line, three and four hours' travel 
from the nearest village (Figure 61). The men, while tending 
these fields, live in miserable little cortalos. The distance to the 
village or any building is so great that seed potatoes are buried for 
the winter in pits near the fields. With so many factors elimi- 
nated or equal on the two slopes, the measurements which Bois- 
sier gave nevertheless show a great discrepancy. There is a dif- 
ference of 2 1 5 meters in the height limit of the vine on the two 
slopes, 150 meters difference for the walnut, 520 meters for 
the olive, and 300 meters for the fig. Rye shows distinct dif- 







Fig. 28. High Fields on the Sunny Slopes, Val D'Isere, French Alps 



HEIGHT LIMITS OF FIELDS 



93 



ferences. If insolation is the sole factor, why should the figures 
of differences vary so greatly? The truth is that crops vary in 
sensitiveness to the climatic elements at different numerical 
points of intensity. It may not be insolation which is the deter- 
mining factor of the limit of a particular crop, but the sum total 
of climatic aspects. Moreover, physical conditions are never uni- 
form over any great area. Boissier himself notes that the high rye 
fields on the south side were in certain well protected nooks. One 
is led to suspect the importance of any figures denoting extreme 
limits of crops. It may be that to measure sunlight effect we shall 
be forced to disregard crop limits and crop characters, since so 
many other factors intrude themselves there, as, for example, 
geologic and soil factors. 

The Geologic Factor 

Geology makes itself felt in influencing height limits not only 
as it influences the character of topography but also because it 
has consequences on the soil and influences the capacity of the 
soil for ground water. Sieger, in discussing alp huts, shows that 
their elevation is not so much a matter of exposure as of geology. 
The geology by its determination of soil character, mass of moun- 
tain, and ground water outweighs other physical factors. As the 
alp hut bears direct relation to the alp, the elevation of the hut is a 
climatic and topographic matter. 

Height of Alp Huts in Austria 



Region 


Exposure 


Limit 


Nature of Rocks 


Imst, Zirl 


Adret 


161 1 m. 


Calcareous 




Ubac 


1707 


Crystalline 


Gailtal 


Adret 


i5°9 


Calcareous 




Ubac 


1718 


Schists 


Haut-Pustertal 


Adret 


Q006 


Crystalline 




Ubac 


1766 


Calcareous 


Haute-Drave 


Adret 


1966 


Crystalline 




Ubac 


1701 


Calcareous 



Crystalline massifs generally give gentler slopes, greater mass of 
mountain, and more abundant ground water than do calcareous 
mountains. In the preceding table it will be noted that the mean 
altitude limit on the adret is 1886 meters and on the ubac is 1720 
meters. But let us arrange the table in another form. Then it 



94 MOUNTAIN GEOGRAPHY 

will be seen that the adret calcareous stations are lower than the 
ubac crystalline stations. 

Adret Ubac 

1 6 1 1 crystalline 1707 crystalline 

1 509 calcareous 1718 schists 

2006 crystalline 1 766 calcareous 

1 966 crystalline 1 701 calcareous 

The matter of soil patently influences not only the height of 
agriculture but also its character. A calcareous soil is warm and 
dry. Crystalline rocks create infertile siliceous soils or fertile 
clays. In the Dolomites and Carnic Alps the beech finds its 
highest limits in limestone and its lowest limits in siliceous rocks, 
regardless of exposure. 

The geologic factor influences slope. Great degree of slope, 
besides its more apparent influence on cultures, increases ava- 
lanches and landslips, which locally have more effect in lowering 
cultural lines than at first would be expected. Oversteep slopes 
have no soil. Such slopes commonly have alluvial fans or cones 
at the base. These deposits, though in places subject to torrential 
floods, may be loosely compacted and consequently well drained. 
If the gradient of the erosive portion of the stream's course on the 
fan is not too steep, the deposit may be fine enough for agricul- 
ture. (Figures 28 and 50.) Much depends also on the character 
of the rock suffering erosion. These cones of dejection are favor- 
ite sites for fields and villages. When on the adret they are better 
insolated than on the valley bottom. They are less exposed to 
cold air drainage than the valley, and less subject to floods. 

There is perfect gradation from the alluvial fan to a sheet of 
soil lying on a slope. On a moderately steep slope soil formation 
may not be sufficiently rapid to cover the entire slope in opposi- 
tion to soil slip and soil erosion. In such a case perhaps only the 
lower half of the slope will have soil suitable for agriculture. As 
one passes up a valley, both the stream profile and the crest of the 
flanking mountains increase in altitude. Likewise the limit of 
the fields rises. One immediately suspects that the field limit is 
higher because of greater mass in the interior of the range, but 
quite commonly the increasing field limit is a question of soil. 
The soil maintains its proportional cover of the slope regardless 
of the increasing elevation of the slope. Perhaps the most com- 
mon error in studying height limits is to ascribe to climate a 
cessation of culture really due to thinning of soils. 



HEIGHT LIMITS OF FIELDS 95 

An example of lack of soil on an upper slope was discovered in 
the Vallee d' Ouell near Bagneres-de-Luchon in the central 
French Pyrenees (Figure 27). This is a high side valley with a 
bottom of 1000 to 1400 meters elevation. The flanking mountain 
spurs had an elevation ranging from 1400 to 2100 meters. The 
slope of the valley walls was about 30 degrees. The limit of fields 
was 500 to 600 meters higher at the upper end of the valley than 
near the mouth. Seen from the high vantage point of Super- 
bagneres this upper limit appeared to be a matter of accessibility 
of fields from the villages, or an increase of tillage limits due to 
mass of mountain in the interior of the little massif. Closer ex- 
amination showed that the upper half of the slope did not main- 
tain a sufficiently thick apron of soil. This condition is a common 
one. The soil limit of agriculture undoubtedly has been fre- 
quently mistaken for a temperature limitation. 

Zones in Andorra 

Let us approach the question not from the consideration of the 
isolated factors but by means of regional studies, thus showing the 
interrelation of the factors. We shall consider two regions in the 
Pyrenees: Andorra and the Conflent. The agricultural products 
of Andorra are potatoes, garden vegetables, rye, barley, and 
tobacco (Figures 50 and 51). A few vines exist near Sant Julia, 
the lowest community. Vine culture (the grapes are used today 
for raisins) is now restricted because improved transportation 
makes possible the importation of wines from the Lerida region of 
Spain. 

The cereals of Andorra are largely rye and barley. The people 
eat black bread. Where there are gentle breaks in the valley 
walls and tillable slopes extending up to the level of the summit 
plain, rye occupies the top fields. Elsewhere rye is grown on the 
shoulders of the canyon walls. These fields, the highest in An- 
dorra, lie at 1850 meters, and not infrequently are an hour's 
tramp from the nearest house. The isolated position of these 
grain fields is the result of the eager search for sunlight. The 
deep gorges and their alluvial cones may have but a few hours of 
sun during the day. The high fields are invariably up out of the 
gorge on the sunny exposure of the valley. 

The question of sunny slope and shady slope is here, naturally, 
of the utmost importance. Every inch of the solatia {adrei) is culti- 



96 MOUNTAIN GEOGRAPHY 

vated. True, the umbaga (ubac) is farmed, but the value of the 
harvest is much less than that of the solana, and the upper limit 
never so high. Because of the convex form of the slope, the 
greater the altitude on the umbaga, the more numerous the hours 
of sunlight. Also the shade of the mountain to the south may 
cover the lower slope of the solana. But the crops on the shady 
side are commonly so slight or even precarious as to have relative 
unimportance. It may be stated as a principle of mountain 
geography that though the shady slope may be farmed as exten- 
sively as the sunny side, the economy of the two vary greatly. 

The writer did not feel that the climatic limit of agriculture of 
one sort or another had been reached by the Andorrans. Matters 
of accessibility, distance, difficulty of carrying manure to the 
fields and the crops to the granges, and unwillingness to en- 
croach upon pasture lands, as well as unfavorable soils on the 
plas (plateaus), are factors which obviously all play a part. It 
would almost be easier to determine a lower climatic limit for 
grain in the gorges than to designate an upper limit. In the 
steep-sided valleys the duration of sunlight is so slight that the 
season necessary for the maturing of grain is lengthened. In 
places men may sow one grain crop in autumn before the grain of 
that year is harvested in an adjacent field. The land must then 
either lie fallow a year or be put into field crops. 

Tobacco is the most distinctive crop of Andorra. It had its 
start when smuggling was a recognized profession. Climatically, 
the interesting thing in connection with tobacco is that the 
highest cultivated fields in the basins of Andorra are devoted to 
this plant. The highest field in the basin of Andorra lies in the 
side valley of the Entremesaigucs. These canyon fields are 1455 
meters in altitude, the duration of sunlight is short (on September 
15th, with the tobacco still on the stalk, there were but eight 
hours of sunlight), the canyon is cooled by the icy stream and in 
the evenings it suffers with great regularity from cold mountain 
winds. The explanation of this ability of tobacco to grow at this 
extreme height is that these fields have the advantage of being 
close to summer pastures and therefore to manure. 

In the Conflent 

In the eastern French Pyrenees is a region known as the Con- 
flent. Here is an opportunity to study the question of accessi- 
bility. Two examples may be cited from this region of receding 



HEIGHT LIMITS OF FIELDS 97 

culture lines due to difficulty of access and to changing economic 
and social conditions (Figures 46 and 47). The first is that of the 
abandoned Ferme de Randais. This farm is situated on the 
northern slope of the range on a level spot lying on a spur at 1 700 
meters. Generally the mountain flank has a 50 degree slope and 
is covered with forest growth, where, in occasional clearings, 
cattle graze in the summer. Vegetables and grains were in the 
past successfully grown on this farm. Because of steep slopes and 
the large amount of pasture land available, one suspects that the 
basis of the farm's economy was pastoral. The fact remains that 
at this altitude on a reverse or northern slope it was possible to 
raise crops successfully. The height limits of agriculture, as rep- 
resented by the highest fields in the valleys, are then not a mat- 
ter of decrease of temperature with altitude. Long before the 
climatic limit of tillage is attained, topography, accessibility, and 
water for irrigation control the height limits of cultivation. The 
abandoning of the Ferme de Randais was a question of accessi- 
bility. Accessibility in this case is more a social matter than a 
physical one. In terms of the current values of life in France today 
the farm is too remote. 

The Ferme de Randais was an outlier of the agricultural zone. 
In the zone of continuous agriculture which lies in the level valley 
bottom, the finer details of the limits of fields are usually topo- 
graphic, if we bear in mind that topography influences soil char- 
acter, ground water, and the possibility of irrigation. The topog- 
raphy of the foothills which flank the lower valleys is such that 
gentler slopes (35 per cent) 1 are not incapable of agriculture. In 
two ways the per cent of slope influences tillage. Plow land is 
anything up to 40 per cent of slope, and mattock-tilled fields are 
to be seen on much steeper slopes. The critical incline, as found 
upon the foothills, is that which is too steep for good ground water 
conditions. A slope of given incline will have a greater proba- 
bility of verdure if it is situated upon the higher portions of the 
foothills near the rain-forming mountains. 

In this zone in the Conflent there are many abandoned vine 
terraces. Here is an example of changing economy rather than 
climate. The abandoned vine terraces were cultivated in days of 
meager economy, when there were more people and less money 
in the Conflent. Men could then be hired to work among the 
vines for a franc (pre-war) a day, women for seventy-five cen- 

1 A 1 00 per cent slope has a grade of 45 degrees. 



98 MOUNTAIN GEOGRAPHY 

times, where now the laborer asks twenty-five francs, if he can be 
hired at all. Men carried the manure up the slopes in great 
baskets fastened to the back. Likewise they carried the heavy 
grapes down the slope. The most arduous work the valley farmer 
had was on the steep slopes flanking the valley flat. Here the 
matter was not one of distance but of incline. He became unwill- 
ing to struggle up the hillsides, especially in view of the declining 
price of wine. Today with increasing wine prices the terraces are 
coming into use again. Economic pressure is inducing man to 
take the climb. 

The Valley of the Doron 

In 1928 the writer made a study of the valley of the Doron in 
the Tarentaise (French Alps). The Doron is a stream made by 
two branches, the Doron de Champagny and the Doron de 
Pralognan, meeting at Le Villard, and is itself tributary to the 
Iscre at Moutiers. Between Le Villard and Brides-les-Bains the 
valley has for 12.5 kilometers a true east-west direction. At 
Bozel the La Rosiere enters the main stream from the south, 
having a valley of another 12.5 kilometers. The Doron dc Pra- 
lognan is 20 kilometers in length, not counting its Alpine tribu- 
taries, and has a north flow (see Figure 29). 

Economically the valley is pastoral and agricultural. There is 
also an active tourist trade, some iron mining, and a large chemi- 
cal industry. The agriculture on the adret is of vines, grains, and 
hay. On the ubac hay predominates and grains arc less impor- 
tant. The vine there is totally lacking. Vegetables on both 
slopes are largely represented by the kitchen garden, except the 
potato which is a true field crop. There is a distinct agricultural 
zone, but above this are isolated cultivated fields. These fields 
which lie above the true agricultural zone are on the margins of 
the pasture zone. We shall call this margin the mayen alp\ that is, 
the high pasture for May. On the adret mayen alp, grass pre- 
dominates. The fields on the ubac mayen are scarce. One gen- 
erally thinks of villages as lying on the adret. In the main valley 
of the Doron there are twelve villages on the ubac as against seven 
on the adret. It is difficult to determine the relative prosperity of 
the two slopes. It is a question whether insolation is the sole 
factor in either the quality or the limits of agriculture. Rye rep- 
resents the grain of the mayen. The high rye fields are of excel- 



HEIGHT LIMITS OF FIELDS 



99 



lent stand, not meager crops. One wonders if in reality this is the 
greatest altitude for the region at which rye can be successfully 
grown. It is true there is a well marked agricultural zone which 



Mouhers 




Kilometer 

Fio. 29. Limits of Culture in the Valley of the Doron, French Alps 
1. Permanent villages. 2. Temporary villages. 3. Limits of fields. 4. Limits of 



vine. 5. 



Forests. 6. Glaciers. 



might be of climatic significance. But frequently limits of defi- 
nite agricultural zones are to a large extent due to an attempt on 
the part of the farmer to balance the use of his land to the best 
economic purpose. In any alpine valley there is a certain per- 



IOO MOUNTAIN GEOGRAPHY 

centage of land which is most advantageously devoted to tillage, 
to forests, and to pastures. 

The fields about the alp villages of the Doron do not neces- 
sarily represent the climatic height of tillage. The may en alp holds 
fields for several reasons. People remain in the alp villages in the 
Doron a fortnight in the spring to pasture cattle, cut hay, and 
plant a crop, another fortnight in the autumn to cut hay and 
harvest, and the cattle are again in these stables in the winter to 
consume the hay in the grange. Thus there is a valuable storage 
of manure in these villages. Many of the villages are essentially 
hay and manure depots. It is an important fact in the growing of 
crops near the alp villages that the greater the amount of manure 
spread upon the fields, the shorter the growing season needed for 
grain maturity. In this region every adret village had fields close 
by, and few fields were any distance from the villages. It is a 
burden to haul manure a great distance. True, at this altitude 
where air is rare and does not hold heat well, insolation is a 
highly significant factor. Certainly, also, is it true that there are 
few may en fields upon the ubac. But one is led to the conclusion 
that manure as well as sunshine is a factor in the height limits of 
fields on the adret. Therefore we must not take the height limit 
of fields as a measure of insolation. 

Even on the ubac the lesser height limit of fields may not be 
taken as a measure of the poor insolation. Grass grows better on 
the shady alps than on those which are sunburnt. Granting this 
to be true, though it may be possible to grow grains on the ubac 
at the height of the alp villages, it is better economy to grow 
grass. Hence the limit of fields on the ubac may be a matter of 
economic climatology rather than deterministic climatology. 

The principal evidence of glaciation in the main valley of the 
Doron lies in the fact that the valleys of the two tributaries meet- 
ing at Bozel, as well as the valley of Allues, are suspended. The 
course of the main stream is now in a post-glacial gorge. On the 
ubac of this gorge there is forest and grass, the slope being too 
steep for fields. On the adret, where the slope is not precipitous, 
the gorge is devoted to vines; a direct case of the importance of 
insolation. But does the upper limit of the vine represent the 
limit climatically? Does the angle of slope of this gorge side 
represent the most favorable angle of exposure to the sun's rays? 
One is inclined to think not. The curious thing is that the upper 
limits of the vineyard plot are frequently determined by the more 



HEIGHT LIMITS OF FIELDS 101 

gentle slope. The more gentle slope above the gorge can be de- 
voted to field crops and here vines cannot economically compete 
for area. Vines are maintained, generally speaking, only on land 
too steep for other crops. The distribution of the vine is, then, 
not purely a matter of insolation but of declivity also. 

There is, then, a distribution of fields in the valley of the Doron 
according to insolation; but factors of economy as well enter 
into the question. Level bottom land is lacking because of a post- 
glacial gorge. On the adret, in what would normally be the forest 
zone, there are fields wherever the slope is gentle enough. A 
complement of the fields is the upper pasture lands. In places 
the need for pasture restricts the advance of cultivation. Else- 
where fields invade the zone of pasture at a considerable altitude 
because large deposits of manure call for the cultivation of crops. 
The needs of the community and the details of topography are 
more important than climate in the distribution of fields. Too 
much has been written about the importance of a single factor in 
the distribution of tilled fields. In reality the control is not single 
but complicated. 

BIBLIOGRAPHICAL NOTES 

Zones and Factors in the Height Limit of Fields 

Arenes, J. " Etude phytosociologique sur la chaine de la Sainte-Baume 

et la Provence," in Bulletin de la Societe botanique de France, lxxiii 

(1926), pp. 1016-1022; lxxiv (1927), pp. 65-85. 
Ball, John. " The Distribution of Plants on the South Side of the Alps," 

in Transactions of the Linnean Society of London, v, pp. 1 19-227 (1896). 

This and the preceding article are examples of the huge literature of 

the botanical and ecological study of mountains. 
Bates, C. G. Forest Types in the Central Rocky Mountains as Affected by 

Climate and Soil. Washington, 1924. (United States Department of 

Agriculture, bulletin 1233.) An example of modern ecology. 
B6nevent, Ernest. "Le Manival: Etude de cone de dejection," in 

Recueil des travaux de VInstitut de geographic alpine, iii (191 5), pp. 69- 

100. A detail of soil relations. The importance of local conditions 

is shown. 
Blanchard, Raoul. " La limite septentrionale de Polivier dans les Alpes 

franchises," in La geographie, xxii (1910), pp. 225-240, 301-324. A 

valuable study. 
Boissier, Edmond. Voyage botanique dans le Midi de VEspagne pendant 

Vannee 1837. Paris, 1839-45. 2 vols. Referred to in the text. 



102 MOUNTAIN GEOGRAPHY 

Bonnier, Gaston. " Etudes sur la v6g£tation de la vallee de Chamonix 
et de la chaine du Mont Blanc," in Revue generate de botanique, i 
(1889), pp. 28-36, 79-84, 146-154, 204-211. 

Brockmann-Jerosch, Heinrich. Die Vegetation der Schweiz- Zurich, 
1925-29. 4 pts. A writer who always sees clearly the geographic 
factors. 

Brown, W. H. Vegetation of the Philippine Mountains: The Relation be- 
tween the Environment and Physical Types at Different Altitudes. Manila, 
1 91 9. (Manila Bureau of Science, Department of Agriculture and 
Natural Resources, Publication no. 19.) 

Bruckner, Eduard. " Hohengrenzen in der Schweiz," in Naturwissen- 
schajtliche Wochenschrift, xx (1905), pp. 817-825. 

Chard6n, C. E. "Life Zones in th^ Andes of Venezuela," in Bulletin of 
the Pan American Union, lxvii (1933), pp. 620-633. 

Christ, Hermann. Das Pflanzenleben der Schweiz. Zurich, 1882. Trans- 
lated by E. Tieche as La flore de la Suisse et ses origines. Basel, 1883. 
Material on relief, degree of slope, and exposure. 

Clements, F. E. Plant Succession. Washington, 191 6. (Carnegie In- 
stitution, Publication no. 242.) An important American reference. 

Dainelli, Giotto. "Le zone altimetriche del Monte Amiata," in Me- 
morie Geografiche, supplemento alia Rivista Geografica Italiana, iv (1910), 
pp. 292-363. The Italians have been especially concerned with 
studies of altitude. This is a detailed study of an isolated peak 1734 
meters high. 

Evrard, F., and Chermezon, Henri. "La veg6tation de la Haute- 
Tarentaise," in Bulletin de la Societe botanique de France, lxv (191 8), pp. 
153-209. This region is referred to frequently in the text. 

Flahault, Charles. " Essai d'une carte botanique et forestiere de la 
France (feuille de Perpignan)," in Annates de geographie, vi (1897), 
pp. 289-312, with a map. Excellent. 

Fliickiger, Otto. Die obere Grenze der menschlichen Siedelungen in der 
Schweiz- Bern, 1906. A classic. 

Fliickiger, Otto. "Passe und Grenze," in Mitteilungen der Geographisch- 
Ethographischen Gesellschaft in Zurich, xxvii-xxviii (1926-28), pp. 36- 

65- 

Fritzsch, Magnus. " Uber Hohengrenzen in den Ortler-Alpen," in 

Wissenschaftliche Verbffentlichungen des Vereins fiir Erdkunde zu Leipzig, ii 

(1895), pp. 105-292. A classic article. 
Gaussen, Henri. Vegetation de la moitie orientate des Pyrenees. Paris, 1926. 

The work of a master ecologist. 
Gaussen, Henri. "A View from Canigou: Nature and Man in the 

Eastern Pyrenees," in Geographical Review, xxvi (1936), pp. 190- 

204. 
Haret, Michel. "Le paysage alpine carpatique," in Revue de geographic 

alpine, xiv (1926), pp. 617-657. 



HEIGHT LIMITS OF FIELDS 1 03 

Herzog, Theodor. Die Pflanzenwelt der bolivischen Anden und ihres ost lichen 

Vorlandes. Leipzig, 1923. (Die Vegetation der Erde, xv.) 
Hupfer, Paul. "Die Regionen am Atna," in Wissenschaftliche Verbf- 

fentlichungen des Vereins fur Erdkunde zu Leipzig, ii (1895), pp. 293*- 

362.* 
Kashyap, S. R. "The Vegetation of Western Himalaya and Western 

Tibet in Relation to their Climate," in Journal of the Indian Botanical 

Society, iv (1924-25), pp. 3 2 7~334; 
Kerner, Anton von. " Studien iiber die oberen Grenzen der Holzpflan- 

zen in den osterreichischen Alpen," in Der Wald und die Alpenwirt- 

schaft in Qsterreich und Tirol. Berlin, 1908. A classic. 
Koegel, L. "Die Pflanzendecke in ihrcn Beziehungen zu den Formen 

des alpin Hochgebirges," in Ostalpine Formenstudien, pt. 1, no. 5 

(1923), pp. 5-126. Topographic ecology. 
Marinelli, Olinto. "I limiti altimetrici in Comelico," in Memorie 

Geografiche, supplemento alia Rivista Geografica Italiana, i (1907), pp. 9— 

99- 

Mayer, Robert. "Die Vcrbreitung der Kulturflachen in den Ost- 
Alpen und ihre obere Grenzc, geomorphologisch bctrachtet," in 
Geographische Zjitschrift, xxxii (1927), pp. 1 13-138. Excellent. 

Moore, Barrington. "Physiological Requirements of Rocky Mountain 
Trees," in Ecology, v (1924), pp. 298-302. Example of tree require- 
ments. 

Peattie, Roderick. "The Conflent," in Geographical Review, xx (1930), 
pp. 245-257. Referred to in this chapter. See a critique by P. Arbos 
in Revue Geographique des Pyrenees et du Sud-Ouest, i (1930), pp. 505- 
508. 

Peattie, Roderick. "Height Limits of Mountain Economies," in Geo- 
graphical Review, xxi (1931), pp. 415-428. 

Peyre, M. "La vigne en Suisse," in Revue de geographie alpine, x (1922), 

PP- 495-548. 
Piatt, R. S. "Six Farms in the Central Andes," in Geographical Review^ 

xxii (1932), pp. 245-259. 
Quervain, Alfred de. "Die Hebung der atmospharischen Isothermen 

in den Schweizer Alpen und ihre Beziehung zu den Hohengrenzen," 

in Beitrage zur Geophysik, vi (1904), pp. 481-533. 
Ratzel, Friedrich. "Hohengrenzen und Hohengiirtel," in ^eitschrift des 

Deutschen und Osterreichischen Alpenvereins, xx (1889), pp. 102-135. 

A classic and a review of previous work. 
Reishauer, Hermann. "Hohengrenzen der Vegetation in den Stubaier 

Alpen und in der Adamello-Gruppe," in Wissenschaftliche Veroffentli- 

chungen des Vereins fur Erdkunde zu Leipzig, vi (1904), pp. 1-2 10. 
Robbins, W. W. Native Vegetation and Climate of Colorado in their Relation 

to Agriculture, Fort Collins, Colorado, 191 7. (Colorado Agricultural 

College, Experiment Station, Bulletin 224.) 



104 MOUNTAIN GEOGRAPHY 

Roletto, G. B. " Considerazioni Geografiche sulla Distribuzione del 
Castagno nelle Alpi occidental^ " in Bollettino della Reale Societa Geo- 
grafica Italiana, serie vi, iii (1926), pp. 548-557. This makes an ex- 
cellent trilogy with the articles on the vine by Peyre and the olive by 
Blanchard. 

Roletto, G. B. "Les zones de vegetation des Alpes Cottiennes dans 
leurs rapports avec Peconomie pastorale," in Revue de geographie 
alpine, xii (1924), pp. 645-668. 

Scharfetter, Rudolf. "Die Grenzen der Pflanzenvereine," in Z ur ^ eo " 
graphie der deutschen Alpen Robert Sieger gewidmet (Vienna, 1924), pp. 

54-69- 
Schimper, A. F. W. Plant-Geography upon a Physiological Basis, tr. by 

W. R. Fisher. Oxford, 1903. Chapter on Mountains. 
Schindler, F. " Culturregionen und Ackerbau in den Hohen Tauern," 

in ^eitschrifl des Deutschen und Osterreichischen Alpenvereins, xix (1888), 

pp. 73-82. Historically a basic article. 
Schindler, F. " Kulturregionen und Kulturgrenzen in den Oetzthaler 

Alpen," in J^eitschrift des Deutschen und Osterreichischen Alpenvereins, xxi 

(1890), pp. 62-84. 
Schindler, F. "Zur Kulturgeographie der Brennergegend," in Zeit- 

schrift des Deutschen und Osterreichischen Alpenvereins, xxiv (1893), pp. 

1-20. 
Schmoe, F. W. Our Greatest Mountain: A Handbook for Mount Rainier 

National Park. New York, 1925. 
Schroter, Carl. Das PJianzenleben der Alpen, 2. Aufl. Zurich, 1926. 
Shaw, C. H. " Vegetation and Altitude," in The Plant World, xii (1909), 

pp. 63-65. 
Shreve, Forrest. " Conditions indirectly Affecting Vertical Distribu- 
tion on Desert Mountains," in Ecology, iii (1922), pp. 269-274. 
Shreve, Forrest. "The Physical Conditions of a Coastal Mountain 

Range," in Ecology, viii (1927), pp. 398-414. 
Shreve, Forrest. The Vegetation of a Desert Mountain Range as Conditioned 

by Climatic Factors. Washington, 191 5. (Carnegie Institution of 

Washington, Publications, 217.) A most exact study. 
Sieger, Robert. Beitrage zur Geographie der Almen in Osterreich. Graz, 

1925. One of the clearest of books on the Alps. Excellent. 
Sorre, Maximilien. Les Pyrenees mediterraneennes. Paris, 191 3. 
Tate, G. H. H., and Hitchcock, C. B. "The Cerro Duida Region of 

Venezuela," in Geographical Review, xx (1930), pp. 31-52. 
Taylor, W. P. "A Distributional and Ecological Study of Mount 

Rainier, Washington," in Ecology, iii (1922), pp. 214-236. 
Tits, D. A. "Les zones altitudinales de ve*g6tation dans les Pyr6nees 

orientales," in Bulletin de la Societe Royal de Botanique de Belgique, lvii 

(1924), PP- 31-50- 
Townsend, C. H. T. Vertical Life Zones of Northern Peru with Crop 



HEIGHT LIMITS OF FIELDS 105 

Correlations," in Ecology, vii (1926), pp. 440-444. Zones up to 

14,000 feet, or 4220 meters. 
Warming, Eugenius. ecology of Plants. Oxford, 1909. Chap. xxi. 
Wettstein, R. "Die Pflanzenwelt der Alpen," in Die Osterreichischen 

Alpen, ed. by Hans Leitmeier (Leipzig and Vienna, 1928), pp. 124- 

136. 

Exposure to Sun 

Almost every regional study takes up this important question. See 
also references on isolation at end of Chapter I. 

Blanchard, Raoul. " L'habitation en Queyras," in La geographies xix 
909)5 PP- I 5~44> 97~ 110 - Raoul Blanchard, the creator of the 
Institute of Alpine Geography at Grenoble, is one of the most pro- 
lific writers upon mountain geography. 

Blumer, J. C. "A Comparison between Two Mountain Sides," in The 
Plant World, xiii (1910), pp. 134-140. 

Bonaparte, Roland. " L'influence de l'exposition sur le site des villages 
dans le Valais," in La geographie, xi (1905), pp. 212-216. The im- 
portance of this article should not be measured by its length. 

Huttenlocher, Friedrich. Sonnen- und Schattenlage. Oehringen, 
1923. Of second importance for this discussion. 

Levainville, J. "La vallee de Barcelonnette," in Annates de geographie, 
xvi (1907), pp. 223-244. Of first importance for its data and its 
point of view. 

Peattie, Roderick. "La question de Padret et de l'ubac," in Revue de 
geographie alpine, xviii (1930), pp. 175-187. A critique of European 
studies; incorporated in part in this volume. 

Shreve, Forrest. "Soil Temperature as Influenced by Altitude and 
Slope Exposure," in Ecology, v (1924), pp. 128-136. 

Vessereau, Marcelle. "L'adret et l'ubac dans les Alpes occidentales," 
in Annates de geographie, xxx (192 1), pp. 32 1-333. The best article on 
the subject. 

Andorra 

Armet y Ricart, S. Les Vails cP Andorra. Barcelona, 1906. 

Brutails, J. A. La Coutume d'Andorre. Paris, 1904. The great book on 
this little country. 

Carrier, Else Haydon. Water and Grass. London, 1932. Chap. xix. 

Chevalier, Marcel. Andorra. Chambery, 1925. Has a topographic 
map. 

Chevalier, Marcel. El paisatge de Catalunya. Barcelona, 1928. 

Corey, Herbert. "A Unique Republic, where Smuggling is an Indus- 
try," in National Geographic Magazine, xxxiii (1918), pp. 279-299. 

Dalmau de Baquer, Luis. Historia de la Republica de Andorra. Barcelona, 
1849. 



106 MOUNTAIN GEOGRAPHY 

Newman, Bernard. Round about Andorra. London, 1928. 

Peattie, Roderick. "Andorra; A Study in Mountain Geography," in 

Geographical Review, xix (1929), pp. 218-233. 
Peattie, Roderick. " Wanderungen in Andorra," in Der Erdball, iv 

093°)>PP- 287-290. 
Rios Urruti, Fernando de los. Vida e Instituciones del Pueblo de Andorra: 

Una Supervivencia SehoriaL Madrid, 1920. 
Whittlesey, Derwent. "Andorra's Autonomy," in Journal of Modern 

History, vi (1934), PP- i47- J 55- 

The Tarentaise 

Arbos, Philippe. La vie pastorale dans les Alpesfrancaises. Paris, 1922. 

Arbos, Philippe. "La vie pastorale en Tarantaise," in Annates de geo- 
graphie, xxi (1912), pp. 3 2 3~345- 

Blanchard, Raoul. Les Alpesfrancaises. Paris, 1925. 

Blanchard, Raoul. "Comparaison des profils en long des vallees de 
Tarentaise et Mauricnne," in Recueil des travaux de VInstitut de gtogra- 
phie alpine, vi (191 8), pp. 261-331. 

Carrier, Els6 Haydon. Water and Grass. London, 1932. 

Evrard, F., and Cherrnezon, Henri. "La vegetation de la Haute- 
Tarentaise," in Bulletin de la Societe botanique de France, lxv (191 8), pp. 
153-209. 

Gex, Frangois. La plus haute commune de Savoie; Val adhere et la Haute- 
Tarentaise. Chambery, 1922. 

Onde, H. "La transhumance en Maurienne et en Tarentaise," in 
Revue de geographie alpine, xx (1932), pp. 237-251. 

Peattie, Roderick. " Height Limits of Mountain Economies," in Geogra- 
phical Review, xxi (1931), pp. 415-428. 

Peattie, Roderick. "La question de l'adret et de Pubac," in Revue de 
geographie alpine, xviii (1930), pp. 175-187. 

Rey, Franc, ois. U exploitation pastorale dans le department de la Savoie. 
Chambery, 1930. 



CHAPTER V 
FORESTS AND THEIR SIGNIFICANCE 

Tree Line and Forest Line 

THE forest line on mountains is the upper limit of the more or 
less continuous forest zone. The tree line marks the upper 
limit of scattered trees. The forest line approximates an average 
tree line. Where considerable clearing at the upper margin has 
taken place, tree line and forest line may coincide. Once a coin- 
cidence has been established there is little possibility of scattered 
trees encroaching upon the pastures, because of the destruction 
done to saplings by browsing and grazing. There is also to be 
considered the lower limit of forests. This is usually the upper 
limit of fields. This limit is an economic matter, which depends 
upon topography, soil, and ground water. On steep slopes the 
lower limits of the forest stand push downward. On gentle slopes 
field culture repels the lower limit of trees to a greater elevation. 
The extent to which lumbering operations are carried on deter- 
mines in part the elevation of this lower limit. But lumbering to- 
day ordinarily consists in selecting suitable trees wherever found, 
and does not alter the total area of woods or cause the retreat 
of forest margins. In Southwestern America, where forests are 
primeval, there is a lower limit which is climatic. The basis of 
such a limit is, of course, a matter of rainfall, but in the more 
southern portions of the western American mountains there is 
also the determinant of soil temperatures. Bare soil there, un- 
shaded by tree growth, prohibits the downward advance of trees, 
because the rock temperatures, well over 45 degrees C. in the 
sun, kill off the seedlings. 

Is the upper limit of forests climatic? Imhof would define this 
margin as the result of the sum total of climatic factors. This is 
more true of the tree line than of the forest line. De Martonne 
has studied forest distribution in the Carpathians, where in the 
High Tatra there is little question of economic influences. There 
the limitation is more exactly climatic. There are always the 
factors of soil and topography present here, as in a consideration 



108 MOUNTAIN GEOGRAPHY 

of any vegetation limits. Yet although, generally speaking, the 
limits of primeval forests in mountains are climatic, it is worth 
our attention to analyze the relative importance of the several 
factors. 

There is no question in mountain geography concerning which 
writers more thoroughly disagree than the matter of forests. In 
France Mougin disputes the views of Lenoble. German and 
Austrian authorities are Fritsch, Reishauer, Imhof, and Marek. 
Topographic, climatic, and economic factors are, in fact, inter- 
related and seldom independent. Shreve, studying relatively 
simple conditions in our arid Southwest, sees moisture, tempera- 
ture, and light factors as all playing a part. In many studies 
temperature may well be more or less eliminated as a deter- 
minant in forest limitations, because the disheveled and crippled 
appearance of the trees definitely points to wind and evaporation 
as limiting factors. There is a coincidence between forest line 
and zone of maximum precipitation. The economic factor is 
illustrated where forests arc found on the poorer ground or 
steeper slopes which cannot be pastured. 

Forest limits at one place or another clearly show relationship 
to temperature decrease, amount of precipitation, duration of 
snow cover, exposure to wind, ground water, soil, and relief. 
Which of these factors does the following table show? Obviously 
the matter cannot be decided without detailed consideration of 
each exposure on which the observation was made. (Figure 30.) 

Forest Limits and Exposure 

Exposure 
Region South West North East 

Lauterbrunnental (N. side of Alps) .... 1950 m. 1920 m. 1820 m. 1900 m. 

Ortler Alps 21 31 2154 2100 2120 

Stubai Alps 1842 1974 1805 181 5 

The table on the opposite page is from Roletto, writing upon the 
Cottian Alps. The climate being Mediterranean, there is a lower 
limit of forest due to aridity. Exposure, precipitation, and soil, 
as well, perhaps, as economic factors, are here illustrated. 

The figures for the Po valley are low because of the steep slopes 
found there. No figures are given for the nearby Plateau of 
Larches, because that territory is now exclusively reserved for 
pasturage. There is not even fuel enough for shepherds 5 use. 
The lack of evenness in the figures representing the difference 
between the heights on the two slopes shows that factors other 



FORESTS AND THEIR SIGNIFICANCE 



109 



Limits of Conifers in the Cottian Alps 



Valley 


Upper limit 


Exposure Side op- 
of posite to 
upper that of the 
Lower limit limit upper limit 


Difference 


Predominant 
geological formation 


Varaita .... 


2300 m. 


1 000 m. 


NE 


2000 m. 


300 m. 


Calcareous schist 


Po 


1840 


IOOO 


N 


I550 


29O 


Serpentine 


Pellice 


23OO 


1 100 


W 


!550 


750 


Serpentine 


Germanasca 


23OO 


1000 


W 


1000 


1300 


Mica schist 


Cluson 


245O 


1 100 


N 


2200 


25O 


Calcareous schist 


Bardonneche 


2IOO 


1 100 


NE 


2100 


O 


Calcareous schist 


Doire de C€ 














sanne .... 


2350 


1 100 


N 


1900 


450 


Calcareous schist 


Cenischia . . 


2200 


I IOO 


NE 


1800 


4OO 


Mica schist 



than climatic control the elevation of the limits. The more gen- 
eral the figures, the more likely they are to show climatic in- 
fluences. Clearing, avalanches, landslides, steep topography, and 
the like control the details of the forest limits rather than the 
average elevations. The physiographic influences in their effect 
on forest distribution are so patent that they do not require 
emphasis. 




Fig. 30. Factors in Tree Limits in the Val Roseg, Switzerland 

Solid line = Climatic tree line. 
Broken line = Theoretical tree line. 
Fine line = Actual tree line. 

a, lowered by avalanches; b, by talus and rock movement; c, by glaciers and 
moraines; d, by clearing for alp pastures. (After Brockmann-Jerosch.) 



no MOUNTAIN GEOGRAPHY 



Physical Factors 



The climatic factors, on the other hand, need elaboration. 
Wind is the most important factor in controlling tree lines. No- 
where in nature are the destructive environmental factors more 
effective against tree invasion than at the frontiers on mountain 
slopes. The last trees are gnarled and twisted, their branches 
growing on one side of the trunk and the trunks inclined. The 
uppermost trees are dwarfed. This is because the higher the 
location on a given exposure on a peak, the greater, in general, is 
the wind velocity. Isolated peaks have also greater wind veloc- 
ity than protected slopes of the same elevation in the interior 
region of a massif. We have always to interpret the altitudinal 
importance of the wind in terms of relief and isolation of the sta- 
tion under consideration. Isolated peaks of moderate elevation 
are usually barren of trees, even though their altitudes are con- 
siderably below the level of critical temperature and precipita- 
tion. Evaporation thus accounts for the much-studied 'balds' 
of the Great Smokies of North Carolina. 

The characters of the trees in the zone of struggle against wind 
and evaporation are of two types. One type is the elfin wood, 
whose trees have short, gnarled, oblique stems with long serpen- 
tine branches and long roots. Shrubs are often limited to creep- 
ing forms. The low branches, growing out over the earth, are 
confined to the layer of air near to the ground, which has higher 
moisture content and higher temperature. 

The second type of tree on exposed summits is the stunted 
tree. This is a dwarf, perfect in structure and in imitation of form 
of a normal plant. Such dwarfs, though of age comparable with 
full grown trees of lower slopes, may be less than a meter in 
height. A typical example of this dwarfing has been observed by 
the writer on Mount Killington, the second highest peak in the 
Green Mountains of Vermont. The west side of the summit has a 
convex surface, but the east side has a little cirque, with a sheer 
wall to the peak of i o meters cut into the peak. On the west side 
are straight dwarf trees, fully matured, but only half a meter 
high. In the shelter of the cirque trees stand 10 meters tall, their 
summits reaching only as high as the shelter of the rock cliff. 
Though temperature may play a part, the difference must be due 
largely to exposure and evaporation. 

Since, regardless of altitude, the summits of peaks are fre- 



FORESTS AND THEIR SIGNIFICANCE 



I 1 1 



North 
Fore -Alps 



Central Alps 



Rigi Sant/s Auenguter 



CfiurnuJden Qdi/05 



Sils dernma 



South 
Fore -Alps 



Gngna Sottocenen 



























2300 


23 


SO 












^^*^S 


22&Q 


1 >w 










21SO. 






v V 








eooo 


/ 1 


•cozo 




\ \ 










1 








19SO 










1 

i 






\ 




^j;y_ 




s>o 


i 






V 

\ 










/aro 






\ 






/3O0 










\ 
\ 






/ 










\ 






/ 










\ 






/ 
/ 










\ 




1100 


/ 










\ 




4 












\ 


1000 


/ 
















f" 














• 



Mints 

24 OO 



Legend Tree i_im»t 



Mass of Mountain —•—————.— 
Fig. 31. Tree Limit and Mass of Mountain in Austria 

{After Brockmann-Jerosch.) 

quently bare, the highest mountains, other things being equal, 
have the highest tree limits. The following table possibly illus- 
trates the point, though mass of mountain must also be regarded 
as a factor. 

The Highest Mountains Have the Highest Forest Limits 

Mountain limit Forest limit 

Carpathians 2 5°° m - 2000-2100 m. 

Massif Central (France) 1 800 1 400 

Vosges 1 300 1 000 

It must not be thought that only summits are affected ad- 
versely by evaporation. Winds are often factors in keeping alp 
lands free of trees and in determining the xerophytic nature of 
heath and grass cover. Moreover, evaporation without the 



I 12 



MOUNTAIN GEOGRAPHY 



assistance of winds is an effective agent in limiting trees. We 
have pointed out in previous chapters how the potential evapora- 
tion from soils at high levels is increased by the rarity of the air 
and high insolation. Exposure, of course, plays a part both in 
insolation values and in wind effectiveness. 




Fig. 32. Forest Limit and Mountain Mass in Switzerland 

The solid line represents the forest limits and the broken lines the relative mass of 
mountain. {After Brockmann-Jerosch.) 



It is said by some authorities that the main topographic and 
climatic factor in limiting tree and forest lines is the length of the 
growing season. Though most pines require three months free 
from frost, one type of pine is said to live where the frostless sea- 
son is but 67 days. It has been suggested by others that the moun- 
tain forest limit is controlled by the same factors as the Arctic 



FORESTS AND THEIR SIGNIFICANCE 1 13 

forest frontier, because each has an approximately equal July air 
temperature. The Arctic July average is about 10 degrees. The 
winter minima are probably of slight consequence to tree growth. 
Yet, conclusive as these data seem, we should be wrong in think- 
ing that they offer a complete solution. 

Mass of mountain, of course, affects temperature, but trees 
gain greater heights on massive mountains than the increased 
temperature would seem to imply. (Figures 3 1 and 32.) Thus an 
increased mass of mountain in inner Austria is said by Marek to 
have raised a forest line from 151 1 to 2029 meters (518 meters 
difference), whereas the 8.26 degree mean annual isotherm was 
lifted from 1741 to 1890 meters (only 149 meters difference). 
Imhof says mass of mountain as a factor in forest limits outweighs 
exposure. His map of forest lines in Switzerland certainly shows 
high forest lines on the massive Monte Rosa and Weisshorn 
groups. It is true that in two groups whose peaks are of equal 
elevation the group with the lesser dissection has the higher forest 
lines. Considering a cross section of Austria on the 29th meridian 
east, one discovers a decrease of the mass and of forest line. The 
decreases are not, however, of the same rate. Whereas mass ele- 
vations lose 1222 meters, forest lines lose but 556 meters. A most 
common observation on forest lines is the excess of elevation in a 
central massif as compared with the lower and more dissected 
border ranges. Moreover, the greater the mass of mountain, the 
greater the difference between forest line and tree line. Reis- 
hauer gives the following table for six regions in the Stubaier 
Alps: 

Mass of Mountain, Forest Line, and Tree Line 



Height of mass 


Forest limit 


Tree limit 


Differen 


2094 


l8l2 


1866 


54 


2358 


l8lO 


1894 


84 


2520 


185O 


I94O 


90 


2746 


l68l 


1875 


194 


2827 


^o 


2070 


140 


2863 


2037 


2199 


162 



Brockmann-Jerosch in his Baumgrenze und Klimacharakter offers 
perhaps the most searching and complete analysis of climate and 
tree limit. The following table is characteristic of the details he 
has collected. The stations are arranged so as to make a profile 
of the Alps, north to south. 



Mean annual 


Average 




mperature range 


tree limit 


Mean mass 


39-3° 


1750 m. 


800 m. 


42.7 


I880 


180O 


41. 1 


i960 


2150 


40.5 


2IOO 


I900 


35-5 


J950 


800 



114 MOUNTAIN GEOGRAPHY 

Conditions of Tree Limits in Alpine Stations 

Altitude 

Rigi summit 1 787 m. 

Guttanen 1 055 

St. Gotthard 2096 

Airolo 1 141 

Monte Generoso 1610 

It is probable that evaporation, soil, slope, ground water, pre- 
cipitation, and duration of snow cover are more important here 
in limiting the upward advance of forests than temperature de- 
crease. Both tables bring out the fact that though generalities 
may be constructed showing the temperature relations of forest 
lines, they are, in reality, of little value, for two reasons: (1) the 
details of control of temperature affect the details of forest eleva- 
tions and (2) many other climatic factors play inseparable roles. 

De Martonne reports that the forest limit in the Carpathians 
is more or less coincident with the zone of maximum precipita- 
tion. Marek does not find this true for Austria. If there is a 
decrease of precipitation above a given zone, then, if the moun- 
tain range has sufficient altitude, there surely must be a limita- 
tion to tree growth. Undoubtedly this would be true for the 
Himalayas and Andes. One must remember also the additional 
factors active above the zone of maximum precipitation, such as 
greater evaporation due to increased insolation, lack of cloud, 
high winds, steep slopes, and deficient ground water. Generally 
the greater the precipitation the greater the percentage of area in 
forest, at least above the cultivated zone. This is partly true be- 
cause the greater the precipitation the greater the degree of slope 
that will support trees. It is true again because precipitation 
means cloudiness and a consequent lessening of evaporation. 

Ratzel in 1 889 was one of the first to point out the relationship 
between forest line and snow line. The earlier conception was 
that the duration of snow cover restricted the growing season. 
Since it is soil temperature and plant temperature rather than 
air temperature which really determine the growing season in 
high altitudes, the relationship of snow cover to growing season 
is important. Marek, in Austria, judged the average length of 
snow duration at the forest line to be 221 days. Other things 
being equal, the forest line is about 800 meters below the climatic 
snow line. 




Fig. 33. Avalanche Track in the Grisons, Switzerland 



M^-^t-j 




'$&£s'£#3 



mmsm \ 

■-■-■I n s ~. j. -■ ■" 







Fig. 34. Slope Farm in the Forest Zone, Stubaital, Austria 
Note the hay barns on the valley floor and the post-glacial cutting of the stream. 



FORESTS AND THEIR SIGNIFICANCE 115 

Snow cover is important in other ways. It has been pointed 
out to the writer by Transeau, of Ohio State University, that 
trees in the High American Sierras are found on the ridges rather 
than in the ravines and valleys. The windswept exposures, hav- 
ing less depth of snow, are first bared by melting, and so permit 
seedlings to get an early start. A forest cannot be expected to 
reproduce itself over an area which has snow cover until late 
July or early August. Shaw shows conclusively still another 
effect of snow as a limitation on forests. By visiting the Selkirks in 
winter he proved his contention that the snow 'drowns' trees by 
preventing aeration and thus permitting fungous growth at the 
base of the trunk. This accounts for the dead branches of dwarf 
wood near the timber line. He states that alp pastures in the 
Selkirks are treeless because of the depth of snow cover. 

Plant Factors 

The conception of plant zones is further challenged when one 
considers individual kinds of trees. In the Western Alps there are, 
to the casual observer, definite tree zones. Excluding the tree 
flora of the Mediterranean littoral, which extends only up to 300 
meters, there are four tree zones. Many authorities assign alti- 
tudinal limits to these zones. The altitudes which they give may 
be true of the region for which they write, but the conditions are 
so variable in different regions, and on different sides of the same 
mountain, that it is better to leave precise altitudes out of con- 
sideration. It must not be supposed that these zones are every- 
where developed with equal completeness. On the south slopes 
of the Maritime Alps they are crowded together, and only the 
first and third zones are well developed. Many trees shift 
through two zones. Thus, on the south slopes of the Maritime 
Alps, Pinus sylvestris occupies the north cold shaded slopes at 
low altitudes, but as high altitudes are reached, it switches around 
to the south or sunny slopes, leaving the north side to firs and 
larches. 

The truth is that tree species and varieties have characteristic 
requirements as to 

(1) Soil. This varies in physical and chemical nature with the 
geology, slope, and ground water. 

(2) Length of growing season. A matter of great significance, 
which varies first with altitude and secondly with exposure. 



Il6 MOUNTAIN GEOGRAPHY 

(3) Soil temperature. This has been shown to have distinct 
relation to altitude. 

(4) Moisture. A matter of precipitation totals, percentage in 
snow, seasonal distribution, character of storms, ground water, 
and evaporation. 

The soil requirements of trees are many and various. The 
Aleppo pine, laurel, and holly are frequently found on lime soils. 
The Black Cluster pine and larch grow on more acid ground. 
Soil aeration, presence of available salts, particularly nitrates, 
phosphate, and potash, and the amount of ground water in 
greater or less degree control the prosperity of different types of 
trees. A needle-leaf tree and a broad-leaf tree certainly have dif- 
ferent transpiration rates as well as different abilities to carry on 
photosynthesis. The abilities of the two types of trees in regard to 
these functions are for the moment in dispute. In the Maritime 
Alps, of which we are speaking, it is true that the needle-leaf 
trees predominate on exposed and steep slopes. The deciduous 
trees prevail in valleys and in hidden glens. There are, how- 
ever, many other factors which control the distribution of these 
tree types. 1 

The moisture requirements of trees permit of no generalization. 
The pines, for example, differ greatly among themselves. A 
sharp contrast is also presented between the beech and the spruce. 
The beech avoids excessive ground water, but the beech leaf 
needs the atmospheric humidity of the ubac. The spruce requires 
a highly saturated soil, but will tolerate the very dry atmosphere 
of slopes exposed to wind and sun. Among the oaks also, great 
difference in requirements prevails. The common oak (Quercus 
robur) is a typical deciduous tree in its needs, but the holly oak 
will tolerate the arid and almost soilless precipices of the Fore- 
Alps of the Riviera. 

The upward expansion of any species of tree is, then, the result 
of a complexity of factors, and it is impossible to ascribe definite 
altitudinal limitations. Altitudinal modifications of climate are, 
of course, factors in the height limit of any species. But even 
above that tree line, the same plant will appear as a bush to which 
the botanist assigns a variety name. Thus the Swiss Mountain 
Pine, whose limber branches, supporting great quantities of 
snow and ice without breaking, permit it to live at snowy alti- 

1 Data supplied by Donald C. Peattie. 



FORESTS AND THEIR SIGNIFICANCE 1 1 7 

tudes, will, because of the aridity of peak areas, appear at those 
levels as a bush, variety pumilio. Yet higher, it becomes a creep- 
ing shrub, variety Mughus. 

This section is meant to show that in considering tree or forest 
zones one must study not merely the physical factors, but also the 
reactions of the tree- types to those conditions. The situation is 
analogous to the point of view of cultural geography, where one 
takes into account not only the influences of the earth on man but 
particularly man's adaptation of the earth to his purposes. 

Economic Factors 

The demands which increasing populations place upon for- 
ested areas for lumber and for fuel, the expansion of pasture and 
tilled land at the expense of the forest area, and the demand of 
certain industries, especially the manufacture of charcoal for 
forges, have forced back the margins of the forest in mountains as 
elsewhere. Mougin, judging that the word i Savoie ' once meant 
'dark forest,' believes the rate of deforestation to be proportional 
to the increasing population. Encroachment of pasture area 
upon forest certainly has taken place. There has been definite 
clearing of forests to increase alp pasture. The fuel needed, espe- 
cially on those alps where cheese is manufactured, and the bed- 
ding made by cutting young conifers, have lowered the upper 
limit of forests. The sudden ceasing of forest and the commence- 
ment of the grass lands without transition have been cited as evi- 
dence that forest lines have been forced back. Unfortunately for 
this argument, forest lines in the Western Cordillera of North 
America show this same lack of transition, and yet we know that 
there has been little artificial extension of the pastures there. 

Not a little has been written, pro and con, as to the damage 
that cows, sheep, and goats do to forests. The goats certainly do 
the most injury, but in mountains where forests are worth con- 
sidering the goat is not important. Sheep are alleged by some to 
do considerable damage and by others to do no damage. One's 
interest in sheep raising may well color the argument. There is 
the statement that the greater the size of the flocks of sheep, the 
lower the tree line. On the other hand, would it not be true 
that the lower the tree line, and so the greater the pasture area, 
the larger the flocks? It is the writer's observation, as well as 
that of others, that the elevation of forest lines as seen in the Alps 



n8 



MOUNTAIN GEOGRAPHY 



and other European mountains is seldom the result of climatic 
factors alone. The demand for pasture or the demand for wood 
has altered the forest lines. It is only in countries with a shorter 
history of habitation that true climatic lines are found coinciding 
with actual forest limits. 




Fig. 35. Example of Forest Distribution in the French Alps 
The extent to which forests have been cleared from the adret is well shown. 



The lower limit of the forest, where contact is made with agri- 
cultural lands, is almost always raised as increasing populations 
make demands for a larger tilled area. Koegel offers a detailed 
study of a local retreat of forest lines before the demands for farm 
lands. 

The following table for the Sellraintal is given by Reishauer, 
and goes to show that the use of land for crops on the sunny side 
of the valley contracts the forest lines: 



FORESTS AND THEIR SIGNIFICANCE 1 1 9 

Forest Limits in the Val Camoniga 

Average woods Highest woods Average tree Highest tree 

Sunny slope 1752 m. 1910 1822 2015 

Shady slope with side valleys .. . 1859 J 95 *977 2140 

It is interesting to note here that the differences between the 
highest woods is 40 meters, and between the highest trees, 125 
meters, while the difference between the average woods is 107 
meters, and between the average tree limits, 155 meters. Per- 
haps the maximum limit approximates the climatic limit. 

The extent to which forest distribution is due to economic 
rather than climatic factors is shown by the fact that in low val- 
leys, where evaporation is not a significant element, the shady 
or cool sides hold the forest. In the Vintschgau, 67 per cent of 
the forests are on the ubac, and in the Engadine, 60 per cent. In 
many valleys the percentage is even higher, as is illustrated by 
a valley in Queyras where the percentage is 93. This is not, 
however, the original distribution of forests in the valley. The 
sunny slope in the lower valleys is more favorable for trees, but 
has today been denuded of forests and given over to cultivated 
fields. 

There is a debate among geographers, particularly in France, 
as to the extent to which mountains have been deforested. This 
debate is represented in part by the references given in the bib- 
liography at the end of this chapter. The writings by Mougin 
and Lenoble cover the question. Certainly the Alps were no 
less forested in primeval times than today. Certainly there have 
always been topographic and climatic limitations of the tree 
lines. If one goes far enough back in history, larger ice and snow 
fields opposed the advance of trees. The Stone Age and Bronze 
Age were periods when the open spaces above the forest zone 
were occupied by men who preferred the high meadows to the 
terrifying and chilly shades of the forest. These upland clear 
spaces the German refers to as Urweiden. Deposits of stone tools 
have been found at 2700 meters in Valais. 

But was there serious deforestation in, say, Roman times? 
There certainly was in the Apennines, and, even earlier, in the 
mountains of Greece. Though the Mediterranean buildings were 
largely of stone, the ships were of wood. The demands of Phoeni- 
cian commerce aided in the denudation of the Greek mountains. 
The Etruscans made early demands upon the forests of the Apen- 



120 MOUNTAIN GEOGRAPHY 

nines, and they were followed by the Romans. The climate of the 
deforested Mediterranean mountains was not suitable for natural 
reforestation. The Roman demands for timber then drove the 
woodcutter into the Alps. It is possible that logs were floated 
down the Rhone from as far north as Savoy. But it is in the later 
Middle Ages that deforestation of the Pyrenees and Alps is first 
evident. There appear in the archives of those periods numerous 
laws and regulations calculated to prevent undue spoliation of the 
forests. Examples are readily found in the local and regional 
studies, many of which have historical chapters (see end of 
Chapter VII). 

Deforestation in the Alps and Pyrenees was less a result of 
commercial exploitation than of the overpopulation of the 
valleys. In the later Middle Ages, there was little movement 
of population, and isolated communities were compelled to 
provide for their own natural increase. Not only did the in- 
creased numbers require more cleared fields and pastures, they 
also needed a greater supply of fuel. The degree to which over- 
population caused excessive destruction of the forests is shown by 
the amount of destructive soil erosion which took place. Un- 
doubtedly water resources were also lessened. Natural reforesta- 
tion was hindered by the browsing of the herds and droves among 
the seedlings. 

In the study of deforestation and the preservation of mountain 
forests, it is needful to appreciate the peculiar importance of 
forests in mountain regions. For two reasons there are extensive 
forest areas in mountains: much of the slope is too steep for any 
other use, and the zone of maximum rainfall gives the necessary 
water for tree growth. This is particularly important in arid 
regions. The dry states of Arizona and New Mexico, and dry 
countries like Spain, lean heavily upon their mountain forests. 
Mountains of the volcanic type are lacking in coal; the fuel, 
therefore, in such places, is wood. But forests in mountains have 
so many special local functions that they are guarded carefully 
against overcutting for fuel. The living tree is more important 
to the mountaineer than timber or fuel wood. The farmer will 
preserve trees on the borders of his fields and along irrigation 
ditches, which he will trim and so gather an annual crop of 
branches or twigs. These pollard trees, so familiar an aspect of 
European plains landscapes, are found also in mountain valleys, 
particularly in the Mediterranean mountains. The writer has 



FORESTS AND THEIR SIGNIFICANCE 121 

seen a chestnut stump twelve feet in height and seven feet in 
diameter in the Mediterranean Pyrenees. Above the stump was a 
bush of the last year's growth of shoots. Since this bush is cut 
annually, the diameter of the stump testifies to the years in which 
this cropping has been practised. 

Erosion in mountains is severe. One of the best guards against 
erosion is the maintenance of forest stands. Deforestation in the 
French mountains, particularly in the Pyrenees, has caused such 
almost catastrophic soil erosion as to excite national attention to 
the reforestation of the denuded slopes. The same has been true, 
though perhaps to a lesser extent, in every country sharing the 
Alpine territory. Soil erosion in mountain valleys is a double 
calamity. It not only destroys the slopes, but it also causes tor- 
rential fans of coarse detritus to spread over the valley fields and 
clog the channels of the streams. A farm may be laid waste in a 
single night by gravel washed down by the waters of a cloudburst. 
Unfortunately, when once a gully or ravine is begun the erosion 
is difficult to check. As a preventive, wicker or stone dams are 
thrown up across the gully and then trees are planted there. 
Forests maintain soil on steep slopes. Soil without proper cover- 
ing of vegetation has an angle of rest which is much less steep. 
The wide divergence of angle between these two slopes repre- 
sents the possible amount of earth which may be eroded from the 
slope and deposited over the valley fields. 

It has been many times proved that forests and forest soils 
largely prevent immediate run-off of storm waters and so pre- 
serve the ground water table and regulate the flow of streams. 
The article by Tourney referred to in the bibliography sum- 
marizes opinion upon this matter. Those who wish a quantita- 
tive study of the importance of forests to stream flow and flood 
crests should turn to the article by Bates and Henry. These men, 
both experienced in mountain climatology, determined con- 
clusively that on a mountain in a semi-arid region deforested 
land does not hold snow cover so long as forest, while flood 
crests are earlier and more severe than in the forest area. Besides 
the value of constant stream flow in providing an all year round 
water supply for the mountain village or town, deep snow cover 
means to mountain areas irrigation water, water for power, and 
comparative freedom from floods. Mountains are important in 
semi-arid regions because the heights provide water for irriga- 
tion of the valleys and piedmonts. The industrial future of most 



122 MOUNTAIN GEOGRAPHY 

mountain areas depends upon water power. One of the most 
active groups of agents in furthering reforestation has been the 
water power organizations. The danger of floods to inhabitants 
of the confined valleys hardly needs emphasis. There is yet an- 
other significance to mountain forests. The wood provides ma- 
terial for innumerable, often part-time, wood-working industries, 
and thus gives to the mountaineer a seasonal employment. A 
planning of the land utilization of mountains calls for an exten- 
sive area in tree growth. 



BIBLIOGRAPHICAL NOTES 

Forests 

Bates, C. G. Forest Types in the Central Rocky Mountains as Affected by 
Climate and Soil. Washington, 1924. (United States Department of 
Agriculture, Bulletin no. 1233.) Tree requirements. 

Bates, C. G., and Henry, A.J. "Forest and Stream-Flow Experiment 
at Wagon Wheel Gap, Colo.," in Monthly Weather Review, supple- 
ment no. 30 (1928). Difficulty of reforestation in a dry climate. 

Brockmann-Jerosch, Heinrich. Baumgrenze und Klimacharakter. Zurich, 
1 91 9. Perhaps the most authoritative book on the question. 

Buffault, Pierre. Le Brianqonnais Jorestier et pastoral. Paris, 191 3. 

Chevalier, Auguste. "Le deboisement et la degradation du manteau 
vegetal dans les Alpes," in Annates de geographic, xxxii (1923), pp. 

Cowles, H. C. "The Relation of Snow and Ice to Mountain Timber 
Lines," in Annals of the Association of American Geographers, i (191 1), 
p. 106. 

Demontzey, Prosper. Etude sur les travaux de reboisement et de gazonnement 
des montagnes. Paris, 1878. 

Flahault, Charles. " Les limites superieures de la v6g6tation forestiere 
et les prairies pseudo-alpines en France," in Annates forestieres, x 
(1901), pp. 385-401, 417-439. 

Frodin, John. "La limite forestiere alpine et la temperature de Pair," 
in Botaniska Notiser, 1920, pp. 167-176. 

Gannett, Henry. "The Timber Line," in Bulletin of the American Geo- 
graphical Society, xxxi (1899), pp. 1 18-122. 

Gaussen, Henri. "Les forets du pays d'Ossau," in Revue geographique des 
Pyrenees et du Sud-Ouest, ii (1931), pp. 431-447. 

Gaussen, Henri. "Les forets de la Vallee d'Aspe," in Revue geogra- 
phique des Pyrenees et du Sud-Ouest, iii (1932), pp. 5-17. 

Harvey, L. R. H. A Study of the Physiographic Ecology of Mount Ktaadn> 
Maine. Orono, 1903. {University of Maine Studies, no. 5.) 



FORESTS AND THEIR SIGNIFICANCE 123 

Heybrock, Werner. "The Interval between Tree and Pasture Lines 
and the Position of their Extremes," in Geographical Review, xxiv 

(1934), PP- 444-452. 
Imhof, Eduard. "Die Waldgrenze in der Schweiz," in Beitrage zur 

Geophysik, iv (1900), pp. 241-330. A pioneer article and a classic. 
Kerner, Anton von. Der Wald und die Alpenwirtschqft in Osterreich und 

Tirol. Berlin, 1908. 
Koegel, Ludwig. "Von der Alpinen Buchengrenze," in Zeitschrift der 

Gesellschqft fur Erdkunde zu Berlin, 1929, Heft 1-2, pp. 33-35. 
Kiinkele, Theodor. " Der Hochgebirgswald," in ^eitschrift des Deutschen 

Ound sterreichischen Alpenvereins, xli (19 10), pp. 6-17. Excellent for 

details. 
Lavauden, L. "Un example de degradation veg6tale dans les Basses- 

Alpes," in Revue des eaux et forets, lxxii (1934), pp. 33-36. 
Lenoble, Felix. "La 16gende du deboisement des Alpes," in Revue de 

geographie alpine, xi (1923), pp. 5-1 16. One must be cautious in 

accepting his conclusions. 
Lenoble, Felix. "Le valeur economique du reboisement des Alpes 

meridionales," in Revue de geographie alpine, xii (1924), pp. 5-29. 
Marek, Richard. "Beitrage zur Klimatographie der oberen Wald- 
grenze in den Ostalpen," in Petermanns Mitteilungen, lvi, 1 (1910), 

pp. 63-69. Marek is an authority on forests in the Alps. 
Marek, Richard. " Waldgrenzstudien in der osterreichischen Alpen," 

in Mitteilungen der geographischen Gesellschqft in Wien, xlviii (1905), 

pp. 403-425- 
Meylan, R6ne. "La foret du Risoud,' , in Bulletin de la Societe Neuchd- 

teloise de geographie, xxxiv (1925), pp. 5-15. A forest due to inversion 

of temperature. 
Mougin, P. " Les forets de protection en Savoie," in Revue des eaux et 

forets, lii (1913), pp. 545~557- 
Mougin, P. "La question du d6boisement des Alpes," in Revue de 

geographie alpine, xii (1924), pp. 497-545. Disagrees with Lenoble. 
Mougin, P. " Le deboisement des Alpes du Sud," in Revue des eaux et 

forets, lxxii (1934), pp. 194-198. 
Pearson, G. A. "A Meteorological Study of Parks and Timbered 

Areas in the Western Yellow-Pine Forests of Arizona and New 

Mexico," in Monthly Weather Review, xli (1913), pp. 1615-1629. 

The climate of open spaces versus forests. 
Reishauer, Hermann. " Hohengrenzen der Vegetation in den Stubaier 

Alpen und in der Adamello-Gruppe," in Wissenschaftliche Verqffent- 

lichungen des Vereins fur Erdkunde zu Leipzig, vi (Leipzig, 1904), pp. 

1 —2 10. An instructive and valuable work. 
Roman, Joseph. Les causes du deboisement des montagnes d'apres les docu- 
ments historiques du XIII Q au XVIII Q siecle. 1887. 
Salvador, J. "Simples notes sur Pam6nagement et Sexploitation des 



124 MOUNTAIN GEOGRAPHY 

forets pyr£n6enes francaises," in Revue geographique des Pyrenees et du 

Sud-Ouest, i (1930), pp. 58-74. 
Sclafert, Th6rese. "A propos du d6boisement des Alpes du Sud," in 

Annates de geographic, xlii (1933), pp. 266-277, 350-360; xliii (1934), 

pp. 126-145. 
Shaw, C. H. " The Causes of Timber Line on Mountains: The R61e of 

Snow," in The Plant World, xii (1909), pp. 169-181. 
Tourney, J. W. "The Relation of Forests to Stream Flow," in United 

States Department of Agriculture, Yearbook, 1903, pp. 279-288. 
Zon, Raphael. Forests and Water in the Light of Scientific Investigation. 

Washington, 1927. Reprinted, with revised bibliography, from 

Final Report of the National Waterways Commission, 191 2. 



CHAPTER VI 
ALP PASTURES AND ALP ECONOMY 

What Is An Alp? 

ALPS are, in the language of those who live amongst them, 
the grassy slopes above the tree line, the grassy areas in 
hanging valleys, the pastures on the mountain spurs, and the 
steppe vegetation of plateaus and about the peaks. The alp is, 
therefore, not a peak but a mountain pasture. The mountaineer 
has little interest in the peaks as compared with the utilitarian 
alp. He visits the barrens only to shoot game or to look for lost 
sheep. It is but in recent years that the mountaineer has learned 
alpinism from the tourist. 

Sieger, quoting Spann, gives a topographic and economic defi- 
nition. He writes of alps as those pastures in the mountains, at 
least 900 meters above the level of the sea, which offer during the 
favorable season grazing for cattle, and which, though managed 
separately from the valley farms, are an integral part of the 
farmer's economy. He then quotes with approval the definition 
of Wittschieben. Wittschieben speaks of alps as territorial 
establishments which carry cattle during the summer for a long 
and continuous period of grazing. Wittschieben excludes from 
his definition uplands where cattle go only for a short period or 
which are not pastured because of inaccessibility. No doubt such 
fine distinctions have a place in detailed economic studies of alps, 
but to us they are of little concern (Figure 36) . 

There is, in fact, no statement as to the elevation limits of alps 
which will hold universally. Alps are commonly above the tree 
line. In many places the alp zone is in contact with the agri- 
cultural zone. Such may be petites montagnes, and are relatively 
accessible. Grandes montagnes are the true alps and are distant 
from the agricultural village. Spann would refer to the petite 
montagne as an elevated farm pasture in distinction to an alp. 
The difference lies in that the one is managed from the base 
farm, while the alp has a separate administration. A true alp 
should have seasonally inhabited shelters, if not villages. 

The alp is treeless. Botanically it is grass land which grades 



126 



MOUNTAIN GEOGRAPHY 



from steppe to tundra. Not only may the plants have xerophytic 
characteristics, but they are adapted to a short growing season. 
The plants are in many cases light-loving. In the upper reaches 




Fig. 36, Topographic Relations of Alp Pastures 

Note Bussenalp, Ober Steinberg and Hubelalp. These arc summer pastures above 
the level of the inhabited zone. Scale 1 750,000. Contour interval 30 meters. 



the subsoil may be frozen and tundra aridity is a dominant con- 
trol. Even though day temperatures, especially soil tempera- 
tures, are high, it must be remembered that the minima are most 
significant. As one ascends, the plant associations less and less 



ALP PASTURES AND ALP ECONOMY 127 

resemble those of the valley. The grass becomes thinly scattered 
as in deserts. The pioneer lawn plants give way to plant bunches 
or plant cushions. Above, on the almost barren peaks, are rock 
plant associations. One essential difference between the alpine 
climate and an arctic climate of equivalent mean annual tem- 
perature is the great noon insolation of the alp land as compared 
with that of the true tundra. A difference in the plant life results. 
Alpine peaks have many more varieties of plants than arctic 
fields. 

One of the most common of limiting factors in the upward 
expansion of alp pastures is topography. A steep cliff frequently 
marks the approach to a peak. Literally the alp grows upon a 
cone surface of mantle rock which does not extend completely to 
the peak. Because of the structure of the bed rock, the cone 
when found on sedimentary masses has a declivity whose slope is 
interrupted by cliffs or steep areas. These steepened portions 
may be barren, in forest, or in grass. Each step of the pastoral 
ladder which the herds mount is known to the Germans as a 
Staff el, i. e., ladder round. 

If the alp is, in a way, an alluvial cone clinging about the 
peak, the height of the peak is of prime importance. One must 
remember, however, that the height of the peak is only a partial 
measure of the mass of mountain. The greater the mass of moun- 
tain the higher the summit, and, therefore, ordinarily the higher 
the limit of the alp. Also, if the mass is great, the area of the alp 
is correspondingly extensive. This is true because massive moun- 
tains in most cases have convex rather than concave slopes. Mas- 
sive mountains may have an extent of alp land quite out of pro- 
portion to the area of the valley hay lands. Such is true of the 
schist uplands of the Tarentaise. Therefore, in order to utilize 
fully their alp pastures, the peasants must import beasts each 
summer (transhumants) in addition to the animals they are able to 
winter on the valley hay. In contrast there are the limestone 
border Alps, such as the Chartreuse, where mass is less than in the 
Tarentaise and slopes are more concave. Alp pastures there are 
deficient. In the Tarentaise alp pastures make up three-fifths of 
the area. In the Chartreuse such lands are but one-fifth of the 
area. The Stubaier Alps, a part of the core range of Austria, 
have considerable mass. Alp pasture limits there mount to the 
great height of 3000 to 3100 meters, whereas in the adjacent 
limestone border ranges the limit of pastures is 2400 meters. 



128 MOUNTAIN GEOGRAPHY 

Though mass of mountains is an important factor in the height 
limit and area of high pastures, yet topography can counter- 
balance mass as a factor. There are areas of alp which have 
peaks higher than 3000 meters yet alp limits of only 1600 to 2500 
meters. The Italian Dolomites are an example in point, with 
their steep cliffs in the upper reaches of the peaks. 

However, air temperatures are not so significant in limiting alp 
pastures as they are in limiting cultivated fields. In high alti- 
tudes, it must be remembered, low air temperatures are counter- 
acted in part by high soil and plant temperatures. This is well 
illustrated by fine pastures on sunny slopes under the same air 
temperatures as permanent snow fields lying on the shady side. 

Snow cover will, of course, control the length of the growing 
season. Snow cover, it will be remembered, is the result of the 
amount of snowfall, exposure to wind, exposure to sun, and posi- 
tion in regard to drifting snow or avalanche, as well as other fac- 
tors. An area which is free from snow but a month a year cannot 
be expected to maintain grass cover. 

The matter of soil temperatures deserves further remark. Soil 
temperature depends upon the physical character of the soil, the 
direction of exposure, the duration of sunlight on a given expo- 
sure, the angle of slope, and the vegetal covering. Shady slopes 
may have soil temperatures lower than the already low air tem- 
peratures. Soil temperatures on sunny slopes, especially where 
there is protection from wind, may be surprisingly high. 

By way of resume, the following are the more important factors 
in limiting the elevation of alp pastures: 

1 . Mass of mountain. 

2. Topography. 

3. Soil. 

4. Exposure. 

5. Precipitation. 

6. Evaporation. 

There are a number of terms connected with alps and alp ex- 
ploitation with which the student searching through European 
literature should be familiar. As with any specialized industry, 
there has arisen a set of terms characteristic of mountain pastoral 
life. Moreover, the multiplicity of mountain dialects has given 
a variety of forms based often on the same root. A few of the 
more common are here reproduced. 



ALP PASTURES AND ALP ECONOMY 129 

The alp is, of course, the upland pasture. Such has been the 
use of the word throughout this book. The term alp is perhaps 
the most universally used. Some American physiographers have 
selected the term alb for use. The writer sees little reason to use 
other than the widely accepted aim or alp. There are the follow- 
ing equivalents, or partial equivalents: 



alpe 


berg 


galen 


alpo 


berge 


pla 


arpe 


olbe 


planina 


alpage 


monti 


plaroure 


albe 


montagne 


jasse 


aim 


montagna 





In the maps one frequently see descriptive combinations, as 
Sennberg, Kuhberg, or Viehberg. Arbos indicates a few of the 
numerous place names which involve alpine pursuits. They are 
such as Alpettaz, Arpette, Aups. Calmis, an ancient term for pas- 
ture, has given Lachat, Bellachat, la Chalmette, les Chalps, and 
many other words. Another ancient term for pasture is Late, 
Lee, or Lex. We have then L'Allee Blanche, L'Aile-Froide, and 
Vers PAllee. From la Montagne we have le Mont Bas, le Mont 
Froid, etc., etc. There are also La Vacherie, le Col des Genisses, 
le Jas des Agneaux, la Cabane des Mulets. The other languages 
may be expected to have an equal number of variations. The 
first Staff el or more nearly level area of the alp holds a temporary 
dwelling. Here hay is cut, in some cases land is tilled for field 
crops. This is known as the montagnette, Voralp, may en, Vorsass, or 
Mainensass. Following are a series of French terms demonstrat- 
ing the character of alp exploitation: le Col du Fruit (cheese), le 
Fruit Commun, la Vacherie, la Veliere, le Col des Genisses, le 
Roc des Boeufs, la Cabane des Mulets. The following terms de- 
scribe in a collective word portions of alps. L'Aoup grand is a 
large alp and l'Arpilhoun a small alp. The side of a little alp is 
Couesta de PAoupet. A poor alp is l'Aoupenas. A short grass 
pasture, as for sheep, is Pelouniera. One special term is worth 
note: a Kuhstoss is a cow's portion of grass. A Kuhstoss is thought 
of as supporting two heifers, three sheep, four pigs, or eight 
goats. 

The enumeration of terms is partly to give an idea of the multi- 
plicity of provincial dialects, and the reader must remember that 
the lists are not complete, but rather indicative of the variety. 



130 MOUNTAIN GEOGRAPHY 

Within a single dialect there are numerous terms for the shades of 
distinction. This is characteristic of primitive peoples who are 
highly specialized in their economic pursuits. The nomad of the 
steppes has an interminable vocabulary for the shades of horses. 
The French Canadian has a large specialized vocabulary for the 
kinds of ice that clog the mill races in winter. 

Alp Economy 

Certain economic aspects are peculiar to mountain pastoral 
life in distinction to pastoral life on plains. Some of these have to 
do with the relative isolation or difficulty of access of the alp pas- 
ture. Others have to do with qualities inherent in the pastures 
themselves. 

Most alp pastures are more or less within sight of the valley 
settlements in whose economy they are involved. Yet, especially 
in a valley with a decided glacial U-shape, there may be diffi- 
culty of access which isolates the alp in terms of effort and hours 
of climbing so as to affect its economy. 1 Indeed the alp which is 
easily attainable from the valley settlement is thought of in special 
terms and hardly considered a true alp. These alp lands which 
are at no great distance from the winter settlement or even ad- 
jacent to it are often owned in separate parcels and are treated 
like so much pasture land on a plain. In any case they do not 
have the economic organization of the true alp, located high on 
some mountain shoulder or hanging valley. The isolation of 
these true alps is not so much an isolation of distance as difficulty 
of access. This difficulty is not only in the labor of climbing but 
in the problem of road building and maintenance. That the 
arduous and expensive maintenance of a road is demanded, calls 
ordinarily for corporate efforts. Alps are for this reason, as well 
as others, owned ordinarily by communes, corporations, or 
syndicates. 

More important than the difficulty of access are certain quali- 
ties inherent in the alps themselves. Because the cattle enter upon 
the lowest rung or Staff el of the alps while the upper reaches still 
lie under snow it is impossible to grant small areas to private 
ownership. The alp is a land which because of topographic 

1 The accessibility of an alp must be thought of partly in terms of cow travel. 
Even alps near to the permanent village hold the cattle from nightly return to the 
stables because travel impairs the health of the cow and the milk yield. 



ALP PASTURES AND ALP ECONOMY 



131 



climate must be a range land. Also, if any portion of the alp 
suffers erosion because of misuse, areas below may suffer tor- 
rential deposition, and areas above may experience gullying 
from head ward erosion. Alp use implies access to water supply 




<^Vs 



^ £1 



Fig. 37. Ground Plan of Up-Mountain Movements, Murau Region of 

Austria 

Black = permanently inhabited. Diagonal lines — occupied May 15. 
Horizontal lines = occupied June 15. Dots = occupied August 1. 
Blank = unoccupied. {After Spreitzer.) 

and, if cheese is made, to fuel. In short, alps because of their na- 
ture should ordinarily be operated as a unit. Alps may, in fact, 
be owned privately, by corporations, or by the commune. The 
private ownership is not always the most fortunate. The private 
owner may not have money to preserve and care properly for the 



132 



MOUNTAIN GEOGRAPHY 



alp. Moreover, he is less subject to public opinion and may per- 
mit erosion or other deterioration of the alp. The communal 
ownership is by far the best. The economy of the valley village is 
not complete without utilization of the alp pastures. When a 




Fig. 38. Ground Plan of Down-Mountain Movements, Murau Region 

of Austria 

Blank = uninhabited. Fine dots = occupied until September 15. Coarse dots, 
until October 1 . Horizontal lines, until October 1 5. Diagonal lines, until December 
27. Black = permanently inhabited. (After Spreitzer.) 



commune owns the alp a committee decides the dates of pastur- 
ing, the number of cows permitted on the alp, the amount of 
manuring, and improvement of the grass. 

The right to pasture a cow upon an alp is characteristic of this 
system. Such rights are sold or inherited. The right properly 



ALP PASTURES AND ALP ECONOMY 133 

should, and usually does, go with a piece of valley land. That 
land should be at least of a size and soil that will provide winter 
hay for the beast. But less overgrazing and the attendant destruc- 
tive soil erosion come about if the number of cows on an alp is 
strictly limited. An example of pasture regulation is shown in the 
case of the Beaufort alp in Savoy. Here on an alp of 1 50 hectares, 
half of which is barren, there are 150 cattle for 82 days. The alp 
lies between 1500 and 2000 meters. Had the pasture zone been 
higher, the number of beasts might have been decreased. 

There are in most mountain areas, but particularly in the drier 
mountains, two types of mountain pastures. One has more 
ground moisture and lusher grass. This is pasture for cattle. 
Such lands are founcl on gentle slopes, in valley bottoms, and in 
cirques and other hollows. The second type of pasture has less 
ground water, less succulent grass, and is reserved for sheep. It 
is found on steep slopes, at great altitudes, or on summits exposed 
to considerable evaporation. 

If there is an excess of summer cattle pasture, cows are either 
underfed in the winter, or, what is more likely, are sold in the 
autumn. This occasions the autumnal cattle fairs in so many of 
the commercial valleys. Such a fair is that of Chur in Eastern 
Switzerland or of Moutiers in the French Alps. Andorra, with 
an excess of summer pastures, has a series of fairs to sell the stock 
that cannot be wintered there. So important are these fairs that 
great seasonal discrepancies appear in the statistics of animal 
populations. The summer figures are in excess of those of winter. 

This lack of balance between hay supply for winter and sum- 
mer pasture is compensated for in a number of ways other than 
the selling of cattle in the autumn. Pasturage is rented to owners 
of stock that live on the plains, a subject that is discussed later in 
this chapter. 

Again, the Voralp or mayen may be used partly for hay produc- 
tion. Also portions of the alps not easily accessible for beasts, or 
dangerous for them, are devoted to hay. These wild meadows 
may even be so dangerous for men as to call for public restric- 
tions as to their use. Often a deficiency of alp land is made up at 
the expense of the forest area. 

Where, on the other hand, a deficiency of pasture exists, and 
there is excess of hay land, cows may be pastured in the valley 
about the permanent settlement, or a commune or corporation 
may lease or own lands outside its watershed. Indeed, the water 



134 MOUNTAIN GEOGRAPHY 

divide in the pass head or on the mountain crest will often be 
such as not to hinder the wandering of cattle outside the valley 
proper. In places the catde are driven over snow, and even 
glaciers, to gain remote grazing grounds. 

The character of ownership of the alp pastures is sometimes of 
geographic consequence. A petite montagne is ordinarily divided 
into plots privately owned. The accessibility, the usually gentle 
slope, and the contiguity or nearness to the fields of the valley 
bottom, impose none of the conditions which cause the grande 
montagne to be owned by common interests. This private owner- 
ship of pieces of alp pasture is characteristic of the Fore-Alps of 
Savoy and of the Inntal of Austria. The ownership is sometimes 
a communal affair, or it may pertain to a syndicate or corpora- 
tion. There are, of course, alp lands held by a single individual 
who rents cow rights to others. So important is the summering to 
each pastoral-agriculturist of the valley, so complementary are 
the summer pastures to the valley hay fields, that individual 
ownership of alps runs counter to the geographic set-up of the 
valley economy. The common ownership seems to have been 
decreed by nature. 

Mountain Nomadism 

The alps set the scene for a seasonal movement of the cattle and 
the herders. It would not be far amiss, philosophically, to say 
that the alps impose this nomadism upon mountain regions. 
This mountain pastoral nomadism is the periodic seeking of the 
grass of the upland pastures. The valley land furnishes hay for 
the winter stabling and perhaps the field crops for food for the vil- 
lage. But the valley is really subsidiary to the upland pastures. 
Another characteristic of mountain nomadism is that it is ordi- 
narily within a single region. If the vertical distance of the wan- 
dering of mountain nomadism is not greater than the horizontal, 
it is at least more significant. There are cases not a few where 
stock is taken beyond the limits of the basin into neighboring re- 
gions. In one Swiss example, the cattle actually migrate 70 kilo- 
meters and mount 1000 meters, climbing over a glacier. Ordi- 
narily, however, the distant pastures are continuous with the 
home pastures, and the generalization holds that mountain 
nomadism is a wandering within a region. 

Both the nomadism of the plains and that of the mountains 







Fig. 39. The Day of the Cattle Down-drive in Andorra 

Cattle descending to winter stables in late October. (Courtesy of the Geographical 

Reviezv, published by the American Geographical Society of New York.) 



P JbSBIIHk' ? if-' 

mSBmmW '- - 



*3£$IKb1! 






JL*^ JJfi 






IIP i 



yfi 




if 
4£ £ 



****! 



*<** m ■■j~ *m . ._«...* BE 




Fig. 40. Transhumance in Andorra 
Part of the 30,000 sheep that leave Andorra in autumn for the Spanish plains. 



ALP PASTURES AND ALP ECONOMY 135 

have a rhythm, and imply a return to the starting point. Plains 
nomadism ordinarily has an extent of movement much greater 
than that of the mountains. In each case the termini of the an- 
nual wandering are complements. The plains nomad carries a 
tent or collapsible hut. The mountain nomad has stone huts at 
intervals along the short course of his summer migrations. There 
is one group in the high French Alps that has a separate set of 
huts for almost every week of the summer. 

The degree to which people migrate from their so-called per- 
manent or winter villages depends largely upon the topographic 
set-up of the valley. Generally, severely glaciated regions have 
considerable altitudinal differentiation between the valley bot- 
tom and the alp pastures. The simplest form of seasonal rhythm 
developed through mountain nomadism is that of the Char- 
treuse. This massif is a steep-sided portion of the Savoian Fore- 
Alps. Its valleys are considerably higher than the main valley of 
the Isere. The villages and hamlets of the interior of the massif 
are near to, or at, the lower limit of the alp pastures. The sum- 
mer alp area and the hay and stable area of winter support are 
adjacent. Near the Chartreuse is the area of the Belledonne 
range. Here the winter granges and stables are isolated from the 
villages. The granges are at the lower edge of the summer pas- 
tures, where, on the first Staff el, hay is made. The villages are 
much lower. The cattle then winter above the villages. In the 
Central Pyrenees the same type of seasonal movement is ob- 
served. The difference between the Chartreuse and Belledonne 
system is largely topographic. There is a forest zone on the Belle- 
donne between the 1000- and 1800-meter contours. Practically 
the entire population lives below 1000 meters and most of it be- 
low 800 meters. In the winter the granges are visited daily by a 
caretaker. 

Another type of nomadism is discoverable in the Conflent in 
the French Pyrenees. Here there are no high-level summer vil- 
lages. Living in a Mediterranean climate, the cattle in winter not 
only eat hay in the stable, but graze in the fields. In summer, 
accompanied by a herder or several herders, the droves and 
flocks ascend the mountains. The cattle go to the moist pastures 
of the jasse areas in the forest or cirques, and the sheep still higher 
to the pla pastures. 

A complex nomadism is characteristic of the mayen, Voralp, or 
montagnette. As defined before, these three analogous terms mean 



136 



M 



MOUNTAIN GEOGRAPHY 

H jgh Pas t u res __ 



Md^ren_Al£_ 



Valley 



M 



J J A 

CHAMPAGNY 



N 



JjlS!l __ £* 8 *.!i r . 1 . 



M«y_en__Al^_ _ 



Valley 



M A M J J A SO 

SA1NTE-FOY 



N D J 



M 



_H ifl[h_ ^a st _u r es_ 



_Ma)ren_A[p_ 



A/ 



J J A 

MONGICOD 



LtOEMO- 

CattU & Herders - 
Villagers 



N 



Fig. 41. Mountain Nomadism in Three Regions of the French Alps 
Top to bottom: Ghampagny; Sainte-Foy; Mongirod. (After Arbos.) 

the lowest alp, used for grazing in combination with haying and 
perhaps agriculture. This alp is ordinarily occupied by man and 
beast for a period in spring and autumn. It is usually above or 
near the upper forest limit, or, at least, is allied with the true alp, 
the grande montagne, by a topographic separation from the valley 
floor. 

The Tarentaise offers the simplest form of the more compli- 



ALP PASTURES AND ALP ECONOMY 137 

cated migration. Let us consider two types of Tarentaise no- 
madism. The Val de Tignes is a portion of the uppermost valley 
of the Isere. The town of the same name is 1 849 meters above 
sea level. Its highest satellite village is at 1936 meters. The sur- 
rounding peaks are from 1900 to 2272 meters. The town of 
Tignes is practically at the lower limit of alps and is so high as 
to have a critically short snow-free period. Here the beasts leave 
the permanent village for shelters on the montagnette . The family 
abides with them for a period in the spring. Hay is gathered. 
Then the cattle, herders, and cheese-makers go up the moun- 
tain, while the remainder of the population descend to the val- 
ley to make hay. The cattle coming down from the true alp 
in autumn are met again by the population at the montagnette, 
where hay is cut once again. People and cattle then remain at 
the montagnette until the hay in the grange is exhausted. This is 
perhaps until Christmas. All then come down to the valley 
village. 

The Bourg-Saint-Maurice type is more characteristic of other 
regions, and perhaps comes close to an average. Bourg-Saint- 
Maurice is a town of the lower valley. Between May 1 5th and 
June 1 st each family sends its beasts to the montagnette. The house- 
hold — even the children — accompany the beasts. Within a 
month the animals with the herdsmen and shepherds go up to the 
true alps, while the family descends to the valley. During the 
spring and autumnal sojourns hay is cut on the montagnette. For 
a time in autumn the stock eat in the little pastures neighboring 
the montagnette village. Later they eat the hay of the grange. 
People and animals then make their way to the lowland village, 
where during the summer hay and vegetables have been har- 
vested. In all, the beasts are on the lower level some seven 
months. The people, aside from the herdsmen, are absent from 
the permanent village but two one-month periods. 

The Champagny type of nomadism is illustrated by Figure 50. 
From January to early May, people and cattle are in the per- 
manent village. There is an early summer sojourn of some six 
weeks in the montagnette village. During the summer, when the 
cattle are on the true alp, the people are alternately in the valley 
and on the montagnette, taking in harvests, principally hay. Be- 
cause of the several summer visits to the montagnette the autumnal 
visit is barely a fortnight. There is a return to the montagnette 
village for more than a month in midwinter to consume the 



138 MOUNTAIN GEOGRAPHY 

very considerable store of hay and such vegetables as are left 
there. 1 

There are further aspects of alp land and valley that affect 
nomadic movements. Latitude, altitude, steepness of slope, con- 
tinuity of slope, exposure, and percentage of local alp land all 
enter into the question. In some cases, physical conditions are 
such as to require as long a residence on the mountain as in the 
lowland, indeed, perhaps longer, though of broken continuity. 
Arbos recounts two such instances, both in the French Alps. At 
Mont-Aimont from January to late May, and from September to 
November, people and cattle are in the valley village. A few 
workers, and perhaps the very old and very young, remain in these 
villages all summer. Even these are in the may en village for the 
early winter period. The summer period sees the cattle and a 
part of the population on an alp. Between seasons the cattle re- 
main pastured in the vicinity of the mayen village. 2 

A second type is that of Ceillac. During the winter season men 
and beasts stay in the valley village. Spring, summer, and au- 
tumn, men and beasts are above. Part of the population settles 
in the mayen village for the summer, while another part of the 
group goes higher up the mountain with the cattle. Here the 
occupation of the so-called ' permanent' village is of shorter dura- 
tion than that of the 'temporary 5 habitation. 

Even though there has been in recent years a trend from gen- 
eral farming to pastoral pursuits, there is, nevertheless, a decay of 
pastoral life in the European mountain valleys. This is due in 
part to the decrease in the number of available workers, but is 
more largely the result of hydro-electric developments. The 
royalties received for water rights free numbers of the people 
from the necessity of pastoral labors. 

In the French Alps many of the highest granges are today in 
ruins. Thus in the commune of Saint-Christophe, mentioned 
earlier in this book as having a winter duration of snow that is 
Siberian, thirteen out of the twenty-one summer chalets have 
been abandoned. Today but half the folk of the Briangonnais 
ascend the mountains in summer. Elsewhere the proportion is 
even less. Where, as in L'Argentiere, factories exist because of the 

1 Jules Blache has simplified the classification of mountain nomadic movements. 
See Revue de geographie alpine, xxii (1934), pp. 525-531. 

2 Montagnette and mayen are here used interchangeably. The first is more char- 
acteristic of the French Alps and the second of the Swiss. Mayen seems the preferable 
term. 



ALP PASTURES AND ALP ECONOMY 139 

electrical power, the decay of migration is most significant. Val- 
loire, which since the beginning of the nineteenth century has 
lost one-half its population, has abandoned five-sixths of its alp 
pastures. 

In the Val d'Anniviers 

The most extraordinary example of nomadism is found in the 
Val d'Anniviers. The writer, unfortunately, has not had the 
advantage of visiting this valley, but its pastoral industry has 
been often studied. The Val d'Anniviers is pendant above the 
south slope of the Upper Rhone. The valley is drained by the 
Navigenze which rises in the massif of Mont Collon and the Dent 
Blanche. It flows into the Rhone at Sierre. The valley has a 
north-south trend. It is 20 kilometers in length. The valley is 
abruptly set in the massif and yet ' hangs ' at a considerable eleva- 
tion above the gorge of the Rhone. 

The chief village is Zinal. The journey from Sierre, in the deep 
gorge of the Rhone river, to Zinal in the hanging valley requires 
six to seven hours on foot. After an hour and a half of travel from 
Sierre one has mounted a thousand feet along a roadway built 
with great difficulty upon the precipitous side of the Rhone 
valley. Beyond Zinal are yet higher villages — indeed, some of 
the highest in Europe. 

The visitor to the valley is surprised to find that always a part 
of the population is on the move. Month by month the people 
move up and down the mountain slope. Because of the rigors of 
the climate, each halting place requires substantial protection. 
There must be shelter for the family, stable, granary, and cellar. 
A plentiful wood supply makes the multiplication of barns and 
outhouses easy, and leads to an exaggerated estimate of the popu- 
lation. 

This valley, of all valleys of the Alps, is noteworthy for its 
seasonal migrations. With the change of the seasons there is to be 
seen the movement of the people impelled by the climate. Up 
and down the mountain go processions, moving from the mayen 
village to the valley village, and from the valley village to the 
village among the vineyards far below in the bottom of the 
Rhone valley. In addition to this migration, there is the move- 
ment of cattle and herdsmen to the high alp pastures. The pro- 
cession is a picturesque sight. The migration from the five vil- 



140 



MOUNTAIN GEOGRAPHY 



lages of the Rhone valley to the high Val d'Anniviers is taken 
together. The priest and mayor of each village lead the com- 
panies. The women and children follow on foot. Behind come 
the crowding cattle, goats, and sheep, driven by an ancient shep- 
herd. With the family are carried the household utensils and 
impedimenta. 

upper Alps JFMA^M J ^A, S O ND 

-to 2760 m 

Highest Chalet -2665 
ee 00 Merces 



Lower Alps 
Winter 
/ 2 20 -1936 m 

















ifijsturcsi 






















1 
I 

1 
1 
























i 
1 
1 














































r-nJ— 
























1 
1 

! 














Winter 








Grain &. • 












Winter 


Resi- 
dence 




Gardens 




Residence 




























Sow S 


jmmer < 


jrains 






H 


r* 


est Cereals ! 












Cut Ha 


y 




""Cut Hay 


































Tend 
VAne- 
yard5 








iar- 
vest 




















in 












Harvest 






S 


ow 6 n 


G 


rain 

1 


Vines 



Valley 0/ Phone 
340 m 

4( 

Fig. 42. Seasonal Movements in the Val D'Anniviers, Switzerland 

Let us recount month by month the typical movements of a 
village (Figure 42). February finds the inhabitants at the prin- 
cipal village. It would be wrong to call this the winter village, 
for they are not here most of the winter. Nor is it the permanent 
village when it is occupied but little more than four months of the 
year. These villages are elevated 1220 to 1936 meters above sea 
level. At the end of February the vineyards of the Rhone valley 
are free from snow. The people of the high valley own lands in 
the main valley. The villagers then descend a thousand meters 
to tend the grapes and to sow some crops. The end of March sees 
the groups mounting again to the chief villages — shall we call 
them datum villages? The end of April means an ascent to the 
mayen villages, where hay is gathered and cattle are pastured 



ALP PASTURES AND ALP ECONOMY 141 

near by. Nightly the cattle are collected so that manure may be 
accumulated. Here people and cattle rest for perhaps seven 
weeks. In late June the herders and cattle, as well as the cheese 
makers, mount to the true alp. The villagers descend. We must 
follow their separate fortunes. 

The cattle and their attendants mount by stages (1800 meters 
and 2780 meters) to the final and highest alp at 2800 meters. 
The most elevated cabane of the herders is at 2665 meters. The 
mountain pasturage begins earlier and ends later than in most 
Swiss valleys. It may last, indeed, for 100 days. Towards the 
end of September the cattle descend, not to the mayen, but to the 
datum village. What has been happening to the villagers during 
the summer? When they parted from the cattle on the mayen in 
late June, they descended past the datum village to the Rhone 
village. Here for a month they were busy with the harvest of 
grains and vegetables. By late July they mounted to the datum 
village in the side valley, where they gathered the hay harvest. 
October and part of November find the villagers again by the 
Rhone working among the grapes. Late in November, villagers 
and cattle are again found in the mayen villages, where the cattle 
are consuming the mayen hay crop. 

This is one of the most complicated of nomadisms. 1 The full 
horizontal range of movement may amount to 20 kilometers. 
The altitudinal range amounts to almost 2300 meters. There are 
three villages for each group. The villagers have fewer resting 
periods than the herders, yet the villagers make eight separate 
moves annually. 

So completely are the special resources used that there is a 
high density of population. In 1900 the valley held 2238 people. 
And it must be remembered that many of these people had three 
or four houses. Generally speaking, in the valley itself there are 
not definite towns, but groupings of houses. The people migrate 
to such an extent that the commune has not the complete func- 
tioning of a plains commune. The commune is merely a political 
unit. Though the cure is a functionary of the political unit, the 
schoolmaster is of the settlement or village. The distinction dates 
back to feudal days and the division between temporal and 
spiritual power, but its preservation is due to the unimportance 
of place in a nomadic regime. 

1 The movements of the people of Chandolin of the same valley are so compli- 
cated as to be quite confusing in description or diagram. 



142 MOUNTAIN GEOGRAPHY 

Pastoral Buildings of the Central Pyrenees 

A type of study which is frequently made is of the variety of 
form and purpose of the dwellings involved in nomadic life. A 
number of these are to be found in the bibliographies. The writer 
offers some notes upon a partial investigation in the French 
Pyrenees. 

The number and distribution of buildings of the gaves of the 
Central Pyrenees of France is misleading unless interpreted in the 
light of the seasonal movements of the people. The problem of 
the houses of these gaves is increased because of the difficulty of 
classifying their uses. There are 'permanent' and 'temporary' 
dwellings, villages, stables, isolated stables, and various types of 
hay barns (fenils). Their uses are many, in that they fit into the 
seasonal rhythm of labor required of the people by the topogra- 
phy of the countryside. In these valleys, generally speaking, there 
is little or no separation between the hay meadows of the valley 
and the pastures of the high slopes. What, under natural con- 
ditions, would have been the forest zone is now woodlots and 
little meadows, each with its hay barn. The summer hamlets are 
within an hour's, or even half hour's, walk, of the winter villages. 
Winter stables for beasts are in many cases in the meadows or at 
the edge of the alp pastures. Some are within ten minutes of the 
village. They are near the hay field. This is to save the hauling of 
the hay — in some cases down steep slopes. Their absence from 
the village means better sanitation there. 

In the Conflent the cattle are pastured on the alp or jasse in the 
summer time and are quite separated from the permanent village. 
In the winter the beasts are stabled in the valley or allowed to feed 
in the meadows. In the Central Pyrenees a different system pre- 
vails. The alp pastures are nearly continuous with the valley 
lands. Moreover, heavy snows in the valleys prevent any winter 
pasturage on the hay meadows. Cold weather, snows out of sea- 
son, or overgrazing frequently force the herds from the upper 
pastures to the lower limit of the alp pastures. Hence there are 
settlements or isolated hay barns and stables near the lower limit 
of the alp pastures or among the meadows of what was once the 
forest zone. Even in summer the cattle may be driven nightly to a 
rude settlement at the lower limit of the alps. These groups of 
huts are known variously as cortals, hordes, or pardinas. They are 
ordinarily merely herders' refuges, but some patches of land are 



sill* 





>.:'-fej 



55 (="• 



.": - && 









.... - ■»■ 9 W ' ' ) - ■ ,: - "I 

'■-... • '■.-.- , . >■>■'.- \ 



■V : 



'*.»>:':.,■: 





— ■ »—■ 



■^^^^^v-i©' 



Fig. 44. Cheese Huts of mi. Marji i.i.n Ai.r, Bernese Oberland, Switzerland 



ALP PASTURES AND ALP ECONOMY 143 

on occasion devoted to vegetables. The writer has visited one, 
far above the tree line and in a region lacking fissile stone, which 
was shingled with squared cow-dung. There is frequently a stone 
corral, into which the cattle are herded so that their droppings 
of the night may be collected. This manure is then accumulated 
and carried in season to the valley fields by cart. This means 
fertilizing of the fields at the expense of the pastures. Were the 
beasts stabled in the villages in the winter these manure depots 
would probably not exist. Such a depot exists in the Vallee de 
Griff where one turns to the trail for the Pic du Midi. 

To illustrate more clearly the seasonal movements of the re- 
gion, let us consider the Bareges valley. It is by this valley that 
one mounts from Luz to the Col de Tourmclet. Along the valley 
lie the villages of Esterre, Viella, Betpouey-Bareges, and Bareges. 
The altitude of Luz is 660 meters and of Bareges 1232 meters. 
One slope town, Sers, lies at 1340 meters. Bareges is a large 
town for its altitude, but few of its people live from agriculture or 
herding. The town is a tourist center, has mineral springs, and 
boasts a military hospital. Above Bareges are some chalets and 
barns and stables occupied in the winter. In summer the stables 
are deserted. The cattle are high on the mountain pastures under 
the care of a community vacher, while the sheep and their bergeres 
wander among the yet higher and more remote pastures. 

The summer villages at the edge of the alp are the equivalent 
of the mayen villages of the Swiss, but they are less well organized 
and of a less concentrated form. The family comes to them in 
May before the cattle have left. Commencing in June, four crops 
of grass are harvested. By September 1 5th, snow has fallen on 
the higher alps and the cattle descend to their winter barns. The 
families and the ever-present pig descend to the valley villages. 
Milch cows are taken to the valleys. The Pyrenees are more 
noted for beef production than cheese. The comparative ab- 
sence of the cheese industry explains the primitive form of the 
hordes as compared with the fruitier es or Sennhutten of the Alps. 

In winter the cattle and sheep are stabled, ordinarily, above 
the winter village level. These stables are the hay barns in the 
forest zone or in the summer village zone. The stable is a dark, 
almost hermetically sealed building situated at a spring or over a 
stream. The beasts are allowed the freedom of the building. 
They seek their water at the trough or the hay at the feeding 
rack. The nearer barns are visited at intervals to renew the hay. 



144 MOUNTAIN GEOGRAPHY 

In some, the more isolated stables, high on the mountain, and 
actually buried in snow, there lives a miserable guardian of the 
beasts. He sleeps on straw, as do the animals. Amid the dark- 
ness and stench of the stable he spends his winter in silence except 
for the occasional lowings of the kine. 

Transhumance: The French Alps 

Though the term transhumance is variously defined, there is 
ordinarily a distinction made between it and nomadism. Trans- 
humance strictly is merely nomadism. When applied to moun- 
tains, the term implies an exchange between the pastures of 
plains and the high pastures of mountains. The beasts find winter 
forage on the plains and summer on the alp lands. Frequently 
the animals are sheep with a sprinkling of goats. In transhu- 
mance, as distinct from nomadism, the pastures of the two seasons 
are remote from each other. Ordinarily in European mountain 
transhumance today the families do not take part in the move- 
ment, but only the sheep and the herders. 1 This driving of 
beasts from plain to distant alp is best developed in Mediter- 
ranean lands. Where winter rains provide pasturage in that 
season in the Mediterranean area, in summer the pastures are 
burnt and dusty, the supply of drinking water is scarce, and the 
heat too intense for the well-being of the animals. Mountain 
pastures are looked upon as a necessary complement to winter 
forage. 

Therefore Spain, Southern France, Italy, the Balkans, and 
the Southern Carpathians are all lands where transhumance has 
a long history and a modern phase. The sheep from Southern 
France move seasonally by the thousands to the Massif Central 
and to the Alps of Haute-Savoie and Dauphine. The movement 
of animals from the delta of the Rhone and the Provencal low- 
lands into Tarentaise and Mauriennc is especially important. 

The monotonously flat lands of the Rhone delta and the stony 
plain of Crau near by are ill fitted for sheep in summer. They are 
dry, the forage is poor, and the heat is bad for the animals. It is 
from these regions that vast numbers of sheep repair annually to 
the alp pastures of Tarentaise and Maurienne. In terms of the 
pastoral industry, the two regions are complementary. There is 
also what has been called an inverse transhumance. Sheep 

1 The Balkans form a notable example. 



ALP PASTURES AND ALP ECONOMY 145 

owners living in the mountains send sheep to winter on the plains 
about the mouths of the Rhone. This is of secondary importance 
as compared with the normal transhumance. 

The means and ways of travel of sheep, herdsmen, and im- 
pedimenta between the two areas of forage were a matter of much 
concern in the Middle Ages and early modern periods. The 
narrow roadways, the lack of provision for rest areas for the 
weary sheep, the tolls at bridges and in towns, the expense of 
feeding and watering led to special sheep-ways being constructed. 
These carra'ires were rights of way, three to seven meters wide, 
which avoided towns and farms. Areas were assigned for rest 
places (relargs) and for the night (pousadous) . Narrow routes 
(drayes) led to the high pastures. The organization of a migration 
of so many sheep, together with the materials of living loaded 
upon donkeys, was not unlike the organization of an army. As 
many as 40,000 head of sheep would make up a single movement. 

Even today large droves of sheep move from Provence and 
Gard towards the highest and least accessible alpages of the 
French Alps. The interesting scene is enacted every year. As 
summer approaches, the advance guard of donkeys burdened 
with the shepherds' impedimenta puts in its appearance. The 
sheep and the goats follow, some of them bearing huge booming 
bells. They literally cover the road. Hundreds succeed hun- 
dreds. Forty thousand are pastured each season in the remote 
Alps of Maurienne and Tarentaise. But the long journey on foot 
is largely a thing of the past. The sheep and goats are driven 
from the Crau and the mouths of the Rhone and the p>lain of the 
Midi to Aries, Pont d'Avignon, and Nimcs, where they are taken 
by rail to the very valleys above which lie the summer pastures 
at such centers as Bourg-Saint-Maurice and Mondane. Modern 
transportation has made the carra'ires superfluous. 

There has been an interesting change in the totals of trans- 
humance since 191 3 in Tarentaise and Maurienne: 

24,000 beasts in 191 3 
41,000 " " 1926 
44,000 " " 1928 
41,000 " " 1930 

This is contrary to the usual course of transhumance today. In 
most regions the practice is decreased as compared to the time 
before the war. Here the figure for 1930 exceeds that of 191 3. 



146 MOUNTAIN GEOGRAPHY 

In Provence, wool and especially meat have increased in price. 
The alps of the high communes of Val dTsere, Saint-Martin-de- 
Belcher, and Tignes have raised their rentals and receive 40 times 
the amount which was paid for pasturage in 191 3. 

Generally speaking, transhumance in Europe is on the decline. 
The large-scale importation of wool from the Americas, Aus- 
tralia, and Asia has offered severe competition. Substitutes for 
wool have lessened the market. In Provence there is now a con- 
siderable trade in fattening African sheep for the market. Be- 
cause less land in Alpine valleys is being used for grain crops, more 
forage is raised for winter feed. Hence the mountaineers make 
greater use of the pastures for their own flocks. Moreover, the 
overgrazing which was a phase of unrestricted transhumance 
created a reaction against the system. Briot, a vigorous critic of 
transhumance, was the chief enemy of this annual visitation of 
hungry animals. Ten years after Briot's attacks, the number of 
beasts visiting alp pastures had been greatly reduced. 

For further details of these movements one should see especially 
Arbos's great work, the article by Onde, the readable book by 
Miss Carrier, and the excellent treatise by Blache. Arbos has 
excellent details on the French Alps, while the survey of Blache 
covers many lands and types. 

In the Balkans 

Nowhere is transhumance of more local importance than in 
the Balkans. The following material is largely condensed from 
Miss Carrier's study on nomadism. The writer has visited Dal- 
matia and Albania casually. 

In Balkan territories the Vlachs are, as a race, nomadic. They 
have clung with remarkable tenacity to their time-honored ways 
of life. From their flocks they derive milk, cheese, and meat for 
food, and skins, wool, and leather for clothing. They practice 
transhumance, except where modern conditions are too restric- 
tive. In a region where the plains in the summer are burnt 
brown, the green pastures of the mountains at that season are 
irresistibly tempting. The seasonal rhythm of movement is from 
steppe to mountain upland. A distinction between Balkan trans- 
humance and that of the Alps lies in the fact that in the former 
type the whole family moves. This, however, is in gradual proc- 
ess of change. 



ALP PASTURES AND ALP ECONOMY 147 

Wallachia is the Rumanian portion of the Danube plain. Be- 
tween this semi-steppe and the flat-topped Transylvanian Alps is 
the chief Balkan transhumance. The Vlachs dwell in piedmont 
villages where life is partly agricultural. In winter the pastures 
of the plain support the sheep. In summer the sheep are pastured 
on the mountains in areas known asplaiouri. Formerly there were 
special ways for passage between these seasonal grazing grounds. 
These sheep roads were called drumul oiior, the equivalent of the 
can aire of Provence. 

The movement involves the entire village. On the mountain 
the summer home is a stina. The way of living in the summer is 
primitive, to say the least, and it is said that a stina can be smelled 
before it is seen. Much of the summer season is taken up with the 
making of sheep's cheese. Swine are raised on whey. Early in 
September the entire group, people, sheep, swine, beasts of bur- 
den, with the children and impedimenta, begin the trek to the 
valley. The flocks have by their milk and cheese supported the 
family during the summer except for such grain as was brought 
for bread. The profit comes from selling the surplus cheese, the 
natural increase of sheep, and the by-product of swine. There is 
also the crop of wool. 

There is a people of Vlach origin who live in the Pindus range 
in Northern Greece. They maintain their proper villages in the 
mountain grazing zone. Since the immediate piedmont of the 
range is settled and not open to them, they travel considerable 
distances to Thessaly, Macedonia, and the lowlands of Albania. 
This is an example of an inverse transhumance, for the true cul- 
ture of these Pindus Vlachs is derived from the upland environ- 
ment. The movement from the plain to the mountain upland is 
begun by the sheep and their herders. The animals clog the 
roads. A traveler may see 60,000 migrant sheep in a day. The 
entire movement from Albania to Greece may involve 3,000,000 
goats and 5,000,000 ewes. Here again transhumance has de- 
clined in the last few decades. Italian interests in Albania are 
creating a market for corn, olive, wine, and citrus fruits. The 
one-time Turkish rule permitted free passage of the flocks. To- 
day custom duties are being charged at the national borders. 

Yugoslavia has several phases of transhumance. A normal 
movement is from valleys of the Adriatic littoral to mountain 
pastures in Bosnia, Herzegovina, and Montenegro, some three to 
six days distant. As part of the littoral is on the islands, the 



148 MOUNTAIN GEOGRAPHY 

writer has seen the last part of the autumnal down drive accom- 
plished in boats, the cattle with their heads over the rail of the 
little sail boat, the sheep crowded beneath, and the goats stand- 
ing adventurously on the fore-deck. Except for Northern Dal- 
matia, this migration does not involve families. 

Winter on the east side of the Dalmatian range is more severe. 
The permanent villages found on the upper levels of this gentle 
Pannonian slope exile their flocks in winter time to the valley of 
the Save in the care of shepherds. At one time there was a swine 
transhumance in Bosnia to let the swine feed in autumn upon the 
acorns of the mountain oak forests. 

There are many other types of Balkan transhumance. Enough 
has been said to show the complexity of types as depending upon 
geographic variations. 

BIBLIOGRAPHICAL NOTES 

Alp Pastures and Alp Economy 

The Bibliographical Notes at the end of Chapter VII will also in- 
clude much material on alp pastures. 

Arbos, Philippe. "L'economie pastorale dans quelques vallees savo- 
yardes," in Recueil des travaux de VInstitut de geographie alpine, i (191 3), 
pp. 45-71. 

Arbos, Philippe. "L'economie pastorale en Suisse," in Recueil des 
travaux de VInstitut de geographie alpine, iv (191 6), pp. 355-363. 

Arbos, Philippe. La vie pastorale dans les Alpes Jrancaises. Paris, 1922. 
Published also as Bulletin de la Societe scientifique de PIstre, xliii (1922), 
716 pp. This is the great book on mountain pastoral life. Bibliog- 
raphy of 400 titles. 

Briot, Felix. Les Alpes jrancaises: Etudes sur Peconomie alpestre. Paris, 
1896. A book which greatly influenced alp management. 

Briot, Felix. Les Alpes Jrancaises: Nouvelles etudes sur Peconomie alpestre. 
Paris, 1907. Factual. 

Briot, Felix. "Economie pastorale de la vallec de I'Ubayc," in Annales 
de la science agronomique jrancaise et etranghe, 3'" ser., iv (1909), pp. 
218-227. 

Cardot, Emile. Veconomie alpestre et la genese du regime pastoral. Besangon, 

1897- 

Cavailles, Henri. "L'economie pastorale dans les Pyrenees," in Revue 

generale des sciences, xvi (1905), pp. 777-783. 
Cavailles, Henri. La vie pastorale et agricole dans les Pyrenees des Gaves de 

PAdour et des Nestes. Paris, 1931 . 



ALP PASTURES AND ALP ECONOMY 149 

D6combaz, Edouard. Veconomie alpestre dans le Canton de Vaud. Lau- 
sanne, 1908. (Statistique des alpages de la Suisse, 7.) 

Flahault, Charles. "Les hauts sommets et la vie vegetale," in La 
montagne, i (1905), pp. 165-184. 

Lefebvre, Theodore. Les modes de vie dans les Pyrenees atlantiques orientales. 
Paris, 1933. 

Maas, Walther. "Die Almwirtschaft in der Ostkarpathen," in £eit- 
schrift der Gesellschqft fur Erdkunde ZM Berlin, 1930, Heft 5-6, pp. 185- 

199- 
Martonne, Emmanuel de. "La vie pastorale et la transhumance dans 

les Karpates meridionales," in Zji Friedrich Ratzels Gedachtnis 

(Leipzig, 1904), pp. 225-245. Excellent. 
Peintinger, Alfred. " Zur Geographie und Statistik der Almen im 

Hochschwabgebietc," in Mitteilungen der Geographischen Gesellschqft in 

Wien, liv (191 1), pp. 324-335. Great detail. 
Pittioni, Richard. " Urzeitliche 'Almwirtschaft,'" in Mitteilungen der 

Geographischen Gesellschqft in Wien, lxxiv (1931), pp. 1 08-1 13. 
Rabot, Charles. " La vie pastorale dans la Savoie septentrionale," in 

La geographie, xxvii (1913), pp. 34 8 ~357- 
Rebsamen, Henri. %iir Anthropogeographie der Urner Alpen. Zurich, 

1919- 
Roletto, G. B. " L'economie pastorale d'une commune du haut Val 

Trompia," in Revue de geographie alpine, viii (1930), pp. 163-174. 
Roletto, G. B. "La zona pastorale delle Valli di Lanzo," in La Geo- 

grafia, ix (1921), pp. 3-25. 
Sieger, Robert. " Almstatistik und Almgeographie," in Mitteilungen des 

Deutschen und Osterreichischen Alpenvereins, xxxiii (1907), pp. 225-226. 
Sieger, Robert. Beitrage zjwr Geographie. der Almen in Osterreich. Graz, 

1925. Defines Aim at length. 
Spann, Joseph. Alpwirtschaft. Freising, 1923. A much quoted work. 

The economics of alp economy in much detail. 
Struby, A. Die Alp- und Weidewirtschaft in der Schweiz> Solothurn, 191 4. 

Very important. 
Struby, A. Die Alpwirtschaft im Kanton Graubiinden. Solothurn, 1909. 
Thallmaycr, R. A. Osterreichs Alpwirtschaft. Vienna, 1907. 

77?£ Val D'Anmviers and Valais 

Berndt, G. Das Val dWnnwiers und das Bassin de Sierre. Gotha, 1882. 

(Petermanns Mitteilungen, Erganzungsheft Nr. 68.) 
Biermann, Charles. " La valine de Conches en Valais," in Bulletin de la 

Societe Vaudoise des sciences naturelles, xliii (1907), pp. 39-175. 
Brunhes, Jean. La geographie humaine, 4*' ed. Paris, 1934. 3 vols. Chap. 

viii. 
Brunhes, Jean, and Girardin, Paul. " Les groupes d'habitations du Val 



150 MOUNTAIN GEOGRAPHY 

d'Anniviers comme types d'etablissements humains," in Annates de 

geographie, xv (1906), pp. 3 2 9~352. 
Carrier, Els6 Haydon. Water and Grass. London, 1932. Chap, xxxiv. 
Courthion, Louis. Le peuple du Valais. Geneva, 1903. 
Desbuissons, L6on. " La vallee de Binn," in La montagne, iv (1908), pp. 

221-230. 
Desbuissons, Leon. La vallee de Binn (Valais). Lausanne, 1909. 
Jegerlehner, Johannes. Das Val tfAnniviers. Bern, 1904. 
Schroter, Carl. Das PJlanzenleben der Alpen, 2. Aufl. Zurich, 1926. 

Special attention to the Val d'Anniviers. 
Wolf, F. O. " Les vallees de Tourtemagne et d'Anniviers," in U Europe 

illustree, nos. 106, 107, 108. 

Movements of Mountain Peoples 

Arbos, Philippe. "Le nomadisme dans les hautes valines savoyardes," 

in La montagne, ix (191 3), pp. 324-340. 
Arbos, Philippe. "La transhumance savoyarde en Provence," in 

Revue de geographie alpine, viii (1920), pp. 665-666. 
Arbos, Philippe. La vie pastorale dans les Alpesfrancaises. Paris, 1922. A 

classic on nomadism and transhumance; exhaustive for the region. 
Blache, Jules. Vhomme et la montagne. Paris, 1933. An excellent re- 
sume, beautifully illustrated, written authoritatively for the general 

public. 
Blache, Jules. "Le types de migrations pastorales montagnardes," in 

Revue de geographie alpine, xxii (1934), pp. 525-531. 
Blad6, J. F. "Essai sur l'histoire de la transhumance dans les Pyrenees 

francaises," in Bulletin de geographie historique et descriptive, 1892, pp. 

301-315. 
Blanchard, Raoul. "Migrations alpines," in Annales de geographie, xxxi 

(1922), pp. 308-312. 
Brunhes, Jean. La geographie humaine, 4* e*d. Paris, 1934. 3 vols. Chap. 

vii. 
Carrier, Else Haydon. Water and Grass: A Study in the Pastoral Economy of 

Southern Europe. London, 1932. One of the most informative books 

on the subject in English. 
Cavailles, Henri. La transhumance Pyreneenne et la circulation des troupeaux 

dans les plaines de Gascogne. Paris, 193 1 . 
Chevalier, Marcel. "La transhumance et la vie pastorale dans les 

vallees d'Andorre," in Revue des Pyrenees, xviii (1906), pp. 604-618. 
Dedijer, Jevto. "La transhumance dans les pays dinariques," in 

Annales de geographie, xxv (191 6), pp. 347-365. 
Fabre, L. A. " L'exode du montagnard et la transhumance du mouton 

en France," in Revue d^economie politique, xxiii (1909), pp. 161-200. 
Fliickiger, Otto. "Die Wanderungen der Berner Bauern," in Mil- 



ALP PASTURES AND ALP ECONOMY 151 

teilungen der geographisch-ethnographischen Gesellschqft in £urich 9 xxi 

(1920), pp. 65-74. 
Fournier, Joseph. " Les chemins de transhumance en Provence et en 

Dauphine," in Bulletin de geographie historique et descriptive, 1900, pp. 

237-262. 
Fribourg, Andre. " La transhumance en Espagne," in Annates de geo- 
graphie, xix (1910), pp. 231-244. A famous article. 
Gasperi, G. B. de. "Le casere del Friuli," in Memorie Geografiche, supple- 

mento alia Rivista Geografica Italiana, viii (1914), pp. 295-461. 
Khristianovich, V. P. Mountainous Ingushiya (Gornaya Ingushiya). Ros- 
tov, 1928. Reviewed by J. V. Fuller in the Geographical Review, xxi 

(1931), pp. 154-155- Included as a study of transhumance in the 

Caucasus. 
Lefebvre, T. "La transhumance dans les Basses-Pyrenees," in Annales 

de geographie, xxxvii (1928), pp. 35-60. 
Lencewicz, Stanislas. "La transhumance dans le val de Rechy," in 

Bulletin de la Societe Neuchateloise de geographie, xxv (19 16), pp. 106- 

121. 
Mouralis, D. " L'emigration alpine en France," in Revue de geographie 

alpine, xi (1923), pp. 223-240. 
Onde, H. "La transhumance en Maurienne et en Tarentaise," in 

Revue de geographie alpine, xx (1932), pp. 237—251. 
Robert-Muller, C., and Allix, Andre. "Un type d'emigration alpine: 

Les colporteurs de POisans," in Revue de geographie alpine, xi (1923), 

pp. 585-634. 
Roletto, G. B. "La transumanza in Piemonte," in Rivista Geografica 

Itahana, xxvii (1920), pp. 1 14-120. 
Sayce, R. U. "An Ethno-Geographical Essay on Basutoland," in Geo- 

graphical Teacher, xii (1924), pp. 266-288. 
Sieger, Robert. " Zur Geographie der zeitweise bewohnten Siedlungen 

in den Alpen," in Geographische Zjitschrift, xiii (1907), pp. 361-369. 
Spreitzer, H. "Der Almnomadismus des Klagenfurter Bcckens," in 

%ur Geographie der deutschen Alpen Robert Sieger gewidnet (Vienna, 1924), 

pp. 70-86. Excellent details. 
Wallner, Hans. " Die jahrliche Verschiebung der Bevolkerung und der 

Siedlungsgrenze durch die Almwirtschaft im Lungau," in MitteiU 

ungen der Geographischen Gesellschaft in JVien, liv (191 1), pp. 358-403. 
Wopfner, Hermann. " Einc siedlungs- und volkskundliche Wanderung 

durch Villgraten," in ^citschnft des Deutschen und Osterreichischen 

Alpenvereins, lxii (193 1 ), pp. 246-276. 



CHAPTER VII 
LAND UTILIZATION AND ECONOMICS 
The Economic Balance in Land Utilization 

MOUNTAIN lands should be studied in three dimensions. 
Previously we have been studying the use of land in the 
dimension of height or depth. Now we shall do well to consider 
those two dimensions which make for horizontal area. This study 
resembles an areal study of a plain countryside, but even here 
there are certain elements which are peculiar to mountains. This 
will be seen when one realizes that the amount of land used and 
the character of the use depends upon morphological and clima- 
tological conditions arising out of the orography. Moreover, 
there is a struggle maintained in mountains to balance the areal 
distribution of alpland, forest, and field because, since one cul- 
tural zone lies above the other, erosive agencies, water supply, 
landslide, and avalanche in one zone may affect the prosperity of 
the zone lower down. 

Rude climate, barren peaks, and the comparative inaccessi- 
bility of large areas may reduce the inhabited territory. In Aus- 
tria, uninhabited areas amount to 54 per cent in the Fore-Alps 
of Salzburg and 36 per cent in the Julian Fore- Alps. The differ- 
ence is largely due to relief. Escarpments in portions of the 
Dolomites cause 85 per cent of the area to be uninhabited. The 
uninhabited zone, the anoekournene, in the Adamello Group, is 87 
per cent of the total, and in the severely glaciated Silvretta Massif 
it is 94 per cent. 

The density of mountain population then should be considered 
as the number of persons divided not by the total area of the 
mountain zone but by the area of the inhabited portion. The 
density of population in mountains considered in this manner 
proves to be very high. This is shown without the need of statis- 
tics by the rapid succession along the valley of agricultural, com- 
mercial, and industrial centers. Generally, in the Alps the in- 
habited land lies between 700 to 2000 meters and the uninhabited 
area is from 30 to 90 per cent. The inhabited and productive 
area maintains a population of 15 to 150 persons to the square 
kilometer. Mountains, if considered in terms of total area, are 
not economically important. On the other hand, there are many 



LAND UTILIZATION AND ECONOMICS 



!53 

Such 



prosperous and populous towns in mountain districts 
towns lie in the productive area. 

Fruh gives the following figures on unproductive areas: Upper 
Rhine area, 36 per cent; Upper Engadine, 51 per cent; Tirol, 25 
per cent. Arbos gives details of land utilization for the entire 




20qq ^ ^ . 
\ 


2691 ^ ' 


1 J) 




* 3116 

t 


1 ~~~" ( 


\ \ J ' 


1 


343*7 " " " 


THE GEO' R I'tur 



Fig. 45. Land Use in the Vai. D'Iserf, French Aeps 
This is a high valley above the true agricultural zone. Figure 37 pictures some of 
the existing fields. (Courtesy of the Geographical Review, published by the American 
Geographical Society of New York.) 

French Alps. Norbert Krebs has a discussion and a map of land 
utilization in his great work on Austria. His article on populated 
and unpopulated areas of the Eastern Alps shows a geologic in- 
fluence upon percentage of unproductive lands in the following 
table: 

Geologic Factors in the Area of Unoccupied Land in the 
Eastern Alps 

Percentage of unoccupied lands 

Gneiss Alps 68 per cent 

Schist Alps 66 

North Limestone Alps 64 

South Limestone Alps 47 



154 MOUNTAIN GEOGRAPHY 

Generally speaking, the percentage of waste land increases as 
one penetrates the mountain chain. Much of the waste land of 
the Bernese Oberland or the Mont Blanc massif is actually under 
snow and ice. The valley of Ubaye has not only the usual sub- 
tractions from total area because of height, slope, and relief, but 
much of the land that might otherwise be productive for field, 
forest, or pasture has infertile soil. About one-half of the valley 
is unproductive because of barrens, scrub lands, and torrential 
areas. 

A large part of the area of mountain valleys is in pasturage and 
hay. Incidentally this percentage devoted to hay, as compared 
with other crops, increases with altitude. Andre Gibert offers the 
following table: 

Hay and Altitude in the Valgaudemar (French Alps) 
(Communes increase in altitude, left to right) 

Saint- Saint- Saint- Villard- CMmcncc- Guillaume- 

Firnun Jacques Maurice Loubiere d'Ambel Pcyrouze 

Per cent of hay, pasture, and 
common land to total sur- 
face 39.6 40.4 63.2 81.6 67.1 34.5 

Per cent of hay, pasture, and 
common land to culti- 
vated surface 122 265 1270 1758 3331 2211 

A chief element in the value of productive land in mountains 
is the ratio of cropped land to pastures. Farm land in Alpine 
mountains has slight value unless adequately accompanied by 
summer pasture area. 

Because of the interplay of economic conditions and the topo- 
graphic and climatic peculiarities of each region, there is at any 
particular time a proper norm for the use of land. How much 
land is best left in field, in forest, and in pasture depends on the 
physical and on the economic factors. There are a great many 
relationships between topography and the use of land which will 
readily suggest themselves. The high communes, for example, 
have often an excess of high sheep pastures, which, in spite of the 
perhaps difficult economy of sheep raising during the winter 
period, can be devoted to nothing else. Field culture on the alp 
pastures may be restricted by the need for pasture. Mountaineers 
have become more and more conscious of the fact that the 
economy of a valley is best served by the preservation of a con- 
siderable extent of forests. 



LAND UTILIZATION AND ECONOMICS 155 

To understand the causes controlling height limits of forests in 
any particular region, one should study the amount of upland 
pastures, the amount of land devoted to crops, and the amount 
devoted to hay. On a farm on a plain, outside the mountains, 
the amount of land left in woodlot is an economic, not a climatic, 
affair. Though the case is not so simple in a mountain valley, 
economic factors nevertheless must here also be taken into con- 
sideration. Because of the commonly prevailing partial com- 
munal ownership in valleys an entire basin may be thought of as 
an economic unit, and adjustments of land use made in the light 
of that consideration. 

How land utilization, altitude, and economy are interrelated 
is best shown by regional descriptions. The writer made a brief 
investigation of the economy of the commune of Naves. This 
commune comprised a hanging valley tributary to the Upper 
Isere in the Tarentaise. The valley has a north-south trend and 
an elevation of 900 to 2413 meters. The difference in elevation 
between the valley of Naves and the main stream is covered by a 
cascade. The commune has four villages and a number of sum- 
mer hamlets. The population in 1928 was 487 people. In 191 1 
there had been 550 persons, and in i860 there were 673 persons. 
Even down to 1900 the commune had had almost complete 
economic independence. The railroad in the valley of the Isere 
did not make its appearance until after 1880, and a good paved 
and well planned road did not greatly precede 1900. 

Since the improvements in transportation, grain culture has 
been on the decrease. The failure of a community to raise its own 
breadstuffs is a measure of the decline of economic independence. 
In 1925-26, only 22 hectares were devoted to the raising of bread 
grains. This would not suffice for half the flour consumed. The 
wheat ordinarily does not grow above 1200 meters. Barley and 
oats are here forage crops. Vegetables are not too common and 
fruits are neglected. Half of the commune is in pasture. Of the 
pasture, at least one-half is high altitude pasture of such inferior 
quality as to serve only for small beasts, that is, chiefly sheep. 
Some hay is cut on fields so isolated as to call for six hay cables to 
transport the hay down steep cliffs to the level of the barns. Four 
of these cables center on the hamlet of Grand-Naves. The longest 
runs two kilometers with a drop of 150 meters. 

Cows are the important beasts. The milk which is the by- 
product of the cheese industry makes swine second in importance. 



156 



MOUNTAIN GEOGRAPHY 



THE CONFLENT 
FRENCH PYRENEES 

SCALE I 63360 



>"* 



" J ~~~ 3 KILOMETERS 

— vjj' Contours m meters 




«>W' vWt^B owraTAlzs V 



VERNCT LES BAINS 




Fig. 46. Chorography of hie Conflent, French Pyrenees, I 
The map is significant only when used in conjunction with Figure 47. One 
hundred per cent slope equals 45 degrees. (Courtesy of the Geographical Review^ 
published by the American Geographical Society of New York.) 



LAND UTILIZATION AND ECONOMICS 



*57 



LAND UTILIZATION 

IN THE CONFLENT 



T WY\ 



Mountain forest 



V^z^^^ l irrigated lands fe^-ffiffiffl Iron mmmg 

P^-^l Foothills ^ 



> 63360 

I MILE 



Green foothill slopes 



Hon- irrigated vineyards 
Hon irrigated grant fields 



2 KILOMCTCKS 



U>..r***-J Mountain heath IM — _g 1 Other non irrigated cultures 

x3 Abandoned terraces 



Uountam grass pasture 




Fig. 47. Chorography of the Confifni, French Pyrenees, II 

Should be read in conjunction with Figure 46. (Courtesy of the Geographical 

Review, published by the American Geographical Society of New York.) 



158 MOUNTAIN GEOGRAPHY 

The high pastures support sheep and goats. The cash income 
depends on (1) dairy products, (2) wood, and (3) live stock. The 
cultivated lands are devoted to wheat, barley, forage crops (bar- 
ley, oats, clover, beets). A surprisingly small area (4 hectares) is 
set out in vegetables for the table. 

Land Utilization in Naves 

Cultivated lands 222 hectares 

Natural meadows 356 " 

Pastures 1 554 " 

Marsh lands 3 " 

Woods and forests 764 u 

Fallow 308 

Other lands 31 " 

3 2 3 8 

An example of economy in a remote and elevated area is found 
in Tavetsch. This area lies at the headwaters of the Rhine in 
Graubunden. It has a topography which has resulted from vio- 
lent forces. Glaciers, torrents, avalanches, and landslides have 
played their parts in making the land forbidding. A high alti- 
tude is accompanied by a rude climate. The high valley provides 
45 hectares of arable land. Most of this is devoted to hay to feed 
the stock, which in summer roams the extensive alpine pastures. 
The highest village is at 1650 meters. Agriculture is precarious 
and grain does not always ripen. Today the growing of grain has 
lost its importance, because transportation by road and rail has 
broken down what was an extreme instance of economic inde- 
pendence. The region continues to have, however, a certain 
degree of isolation. This may explain the fact that, though pos- 
sessed of but meager resources, the region has not suffered de- 
population, but continues to maintain 800 to 900 inhabitants. 

Modern Changes in Land Utilization 

More instructive than these fragmentary examples in the 
utilization of land arc the changes in the use of land which have 
occurred largely in the last half century. It is difficult, however, 
to discuss modern changes in the utilization of mountain land 
without first discussing changes of population and the increased 
facilities of transportation. 

Mountain areas in Europe have suffered severe depopulation 
in the twentieth century. The causes of depopulation, so far as 
they lie in the mountains themselves, are the impoverishment of 



LAND UTILIZATION AND ECONOMICS 159 

agriculture and the decline of pastoral pursuits. Impoverish- 
ment of the soil, soil erosion, and deforestation have been some of 
the causes of decay. An example of impoverishment is found in 
the Val d'Egesse (Conches), which supported 500 cattle in the 
beginning of the fifteenth century. Now but 220 cattle pasture 
there. Many an alp is today barren because of soil erosion or of 
avalanches, landslides, and torrents. Generally speaking, the 
causes of change have been: 

Depopulation. 

Increase of facilities of transportation. 
Increase of agricultural knowledge. 
The development of forage crops. 
Better animal husbandry. 

Mountain regions are difficult to farm. Economically they are 
the last places which are chosen. They are overflow areas of the 
plains, and the number of people that they are compelled to 
maintain depends upon the economic condition and pressure of 
population of the adjacent plain. The height limits of culture in 
mountains in any one century, and almost, one may say, in any 
one decade, is a function of this pressure and of such economic 
matters as price fluctuations. The depopulation of mountain 
areas has increased pastoral pursuits as against agricultural pur- 
suits. When few workers are available, a unit area of land, a 
holding, is operated better as a hay and herding organization 
than as a farm for field crops. The following table is from Arbos: 

Variation of Area in Cereals and its Relation to Population 

Natural regions Cantons Dates 

Baugcs Le Chatelard 1874 

Vercors Villard-de-Lans l ^57 

l 9 l 3 
Baronnics Rosans ^57 

!9H 
Haut-Verdon Castcllane 1857 

1913 
Beaumont Corps ^57 

*9 ! 4 
Tarentaise Bourg Saint-Maurice ... 1874 

!9 ! 3 
Maurienne Lanslebourg ^74 

I91 ? 
Brianconnais Le Monetier 1836 

*9 ! 4 





Grain in 


Hectares 


Pop. 


hectares 


per 1 00 pop. 


10429 


2221 


21 


8029 


I364 


16.9 


5888 


2254 


43-3 


449° 


2048 


46 


3448 


2094 


60 


2381 


JI 59 


48.6 


5617 


2693 


47-7 


3*32 


1663 


53 


5593 


1454 


26 


3673 


1 157 


31.2 


8771 


809 


9-2 


8079 


643 


8 


53*4 


608 


1 1 


4352 


474 


10.4 


5482 


899 


16.1 


3734 


545 


14.6 



160 MOUNTAIN GEOGRAPHY 

The increased facilities of transportation had many far-reach- 
ing results. They brought to the isolated village contacts with the 
easier life of the plains which drew away many of the young 
people. They also brought in the knowledge of better farm 
methods, of which we shall speak. But, most important, they 
made cattle and cheese a better cash crop. The farm had been 
an economic unit where considerable area and labor was de- 
voted to breadstuffs. In the twentieth century the farmer sells 
the 'crop' for which his land is most adapted, that is, dairy prod- 
ucts and beef. The author, visiting a public house in a summer 
village high in the valley of the Schandigg, Grisons, was served 
with bread and cheese. Inquiring about the cheese, he found 
that it was sent in from the Swiss plain, and that the cattle of this 
summer alp were sold as far as Basel for beef. The grain field has 
disappeared. Its area is today devoted to hay meadows. The 
grist mill may still grind cattle feed, or it may stand a picturesque 
ruin beside a broken dam. In the Pays d'Allevard, in the French 
Alps, the cereal area has decreased since 1801 from 528 hectares 
to 81 hectares. Barley and millet, common in the medieval 
mountain agriculture, are disappearing. The periodic fluctua- 
tion of height limits of cereals is economic rather than climatic. 
Flax, a crop of the days of more independence, has disappeared. 
Rapeseed for oil for lighting has been replaced by the introduc- 
tion of electricity. It is significant that the field area in the Tirol 
decreased in some cases 12 per cent in the period from 1875 to 
1922. In Vorarlberg the decrease was 30 per cent. 

The following table from Arbos shows a decided decrease in 
pastoral activities in the last forty years or so for stations in the 
French Alps. It is interesting to note, however, that in several 
cases, as in Bauges, though there is a decrease of cattle, there is an 
increase of the ratio of cattle to population. In other words, the 
population has decreased at a greater rate than the cattle. 

It is clear that the decrease of the fields is not due to an increase 
in the pastures. Pasturage area was decreasing in the same 
period, though to a much smaller extent. The numbers of beasts 
in many mountain regions have declined. Plant rotation, chemi- 
cal fertilizers, and plant selection have increased the crop yield 
per hectare. As cereals on mountain fields were entirely for home 
consumption, fewer hectares were necessary to feed the people 
even had the population remained constant. 

The Lower Tarentaise, French Alps, is an east-west valley 




Fig. 48. Cultivation of a Forty-five Degree Slope, French Pyrenees 
Summer rains in the Central Pyrenees support a heavy grass cover. The grass is 

cut four to six times a season, developing a lawn-like cover which maintains soil 

even on so steep a slope. 

tims mm . wk 






m 









#: 






*! 



1 afc - i 




« Em 



.*tf*d 



as 




Fig. 49. Terraced Farm in Provence, France 

In this unglaciated valley there is no original level land. Farming is possible only 

after terracing. 



LAND UTILIZATION AND ECONOMICS 161 

occupied by the river Isere, between the towns of Moutiers 
and Bourg-Saint-Maurice. 1 The High Tarentaise is the Val de 
Tignes. Down to the end of the nineteenth century Tignes was 
isolated from the rest of the world. The descent in winter was 
dangerous because of avalanches. The climate rendered harvest 
precarious. Snow lay upon the fields in places until June. Even 
in July there were sometimes frosts. 

Relation of the Number of Cattle and Sheep to the Number 
of People in the French Alps 

Total Number Cattle to Sheep to 

Natural regions Canton Dates Cattle Sheep loo people loo people 

Bauges Le Chatelard 1873 7930 1400 75 13 

! 9!3 739 8 595 9 2 7 

Vercors Villard-de-Lans ... 1857 6054 3439 103 58 

1913 4899 731 87 16 

Diois Chatillon l &52 428 2 9°75 6.5 440 

1 91 3 246 21666 5.5 500 

Baronnies La Motte-Chalancon 1852 486 ^549 7 265 

!9 ! 3 85 17575 2.5 500 

Beaumont Corps l8 57 3°^4 10020 55 179 

I 9 I 3 2 485 49*2 64 127 

Champsaur Saint-Bonnet l8 57 6580 25372 56 215 

x 9 ! 3 385° l 3°te 43 *47 

The Lower Tarentaise was more favored for tillage and har- 
vest. It was economically independent. A variety of conditions 
of soil and the comparative isolation made the district a little 
world in itself. Wine, bread, meat, cloth stuffs, and building ma- 
terials all came from the valley. The peasant had all demands 
satisfied. But the essential economy of the valley was determined 
by the need of utilizing the vast upland pastures. Their extent 
was unusually great in comparison to the valley lands. They 
were the real basis of the life of the communes. 

Increased transportation greatly increased the emphasis upon 
pastoral pursuits. Any harvest, as the grain harvest, giving 
mediocre returns, was abandoned. The area devoted to wheat, 
rye, barley, and oats was greatly diminished. After the ravages 
of the phylloxera the vine never regained its importance. A con- 
siderable emigration followed improved transportation. 

So much of the valley land had been devoted to field crops 

1 The following paragraphs are largely based upon a classical article by Philippe 
Arbos, "La vie pastorale en Tarantaise," in Annates de geographte, xxi (19 12), pp. 
323-345- 



1 62 MOUNTAIN GEOGRAPHY 

that there was not enough hay for the beasts in winter. With the 
change in economy, the bringing in of food stuffs and the decrease 
of population, more hay land was available. Cattle raising be- 
came so considerable and so characteristic of the Tarentaise that 
a special breed of cattle, the race tarine, was evolved there. 

The products of the Tarentaise are today almost purely pas- 
toral. Generally speaking, there is an excess of summer pasture 
over hay lands. Many of the beasts cannot be stabled through 
the winter. Therefore in a series of fairs many cattle are sold, 
largely for beef direct to consuming centers. The lambs born in 
September are strong enough to climb to the alps in spring. Not 
so with those born in January. In the spring these are sold for 
butchering. There is accordingly a secondary fair in spring- 
time. 

The summer is a period of considerable cheese production. 
Again during the winter more cheese is manufactured. Butter is 
made also during this period. The necessity of raising young 
cattle to take the place of those sold in the autumn means that 
much milk must be kept for the calves. Thus the economy of the 
region is almost exclusively pastoral. The importance of dairying 
is indicated by the following figures for 1903: 

Tjie Dairy Industry in Tarentaise 

Canton Number of cows Cheese Butter 

Aime 3200 66,000 Kgr. 3580 Kgr. 

Bourg-Saint-Maurice 3 700 147,700 8950 

Motitiers 4900 201,000 45200 

Level Land and Terraces 

An obvious limit to the amount of productive land in moun- 
tains is the degree of slope (Figure 50). Nothing is more valued 
in mountains than level land. In a rough way, level land is a 
measure of the population, other things being equal. If level 
land is subject to floods or covered with torrential deposits it may 
be useless. If high in elevation or too isolated it has less value 
than if at a lower level. But one is impressed in the Western 
American mountains with the fact that every considerable piece 
of level land is or has been the site of a farm. Mountaineers prize 
level land as a dweller in an oasis prizes land with water rights. 
Houses are built on slopes to preserve level land for the plow. 
The piled villages of Tibet are an extreme case of houses built 



LAND UTILIZATION AND ECONOMICS 1 63 

on steep slopes to save the gentler slopes for tillage. So little 
level land is found in some valleys and ravines that in Hoch- 
gallmig in the Tirol there is a saying that " Here the hens have to 
walk on crampons and the cocks use Alpine poles." In Oberinn- 
thal they say, " If the swallows can't find any walls of suitable 
height in the rest of Tirol, they come to Taufers to build their 
nests on the slopes of the valley." In this valley crampons 
(spikes fastened to the sole of the shoe, especially useful in snow) 
have been worn by people going to church. In the Wild- 
Schonau, also in Tirol, it is reported that houses are built on such 
steep slopes that they are moored with chains. In Moos, a village 
not very far from the Brenner, having a population of 800 in- 
habitants, it is reported that more than 300 men and women have 
been killed, since 1 758, by falls from the incredibly steep slopes 
upon which the pasturages of this village are situated. So steep 
are they, in fact, that only goats, and even they not everywhere, 
can be trusted to graze on them, and the hay for the larger cattle 
has to be cut and gathered by the hand of man. These almost 
unbelievable circumstances are, of course, not normal. Neverthe- 
less, that level land is at a premium is attested to by the extent to 
which terracing is resorted to to accommodate the expanding 
populations of the valley bottom. 

Mountains may or may not have level valley land. The young 
valley of a mature mountain group has steep sides and little that 
represents a valley bottom. The mature valley has a floor width 
proportional to the stage of maturity. The glaciated valley of the 
classical U-shape has, of course, level valley land. In this section 
we use the term level for anything approximating a level and not 
classified under distinct slope. The glaciated valley frequently 
has a step profile, so that the level land consists of more or less 
isolated stages, one above the other, along the valley. Moreover, 
unequal deposition in glaciated valleys, definite moraines, and 
the presence of venous 1 will form basins which are either lakes, 
swamps, or fill plains. Both unglaciated and glaciated valleys 
have open places where two streams or two glaciers have joined. 
Such a basin area may be seen in process of formation at the 
broad Concordia Platz of the Aletsch glacier in the Bernese 
Oberland. 

Mountain valleys in maturity normally have natural terraces. 

1 A verrou is a bed rock barrier in a glaciated valley. The writer knows no 
American examples. The origin is not satisfactorily explained. 



1 64 MOUNTAIN GEOGRAPHY 

Without diastrophism or interruption of the cycle of erosion, a 
stream in the processes of meandering and lateral leveling leaves 
terraces. The significance of geologic structure in the formation 
of terraces is an elementary and oft-told tale. 

Valleys have level land and area for farms, for road building, 
and for sites for towns. But there are few cases where level land is 
sufficient to permit even a moderate expansion of the population. 
In lieu of level land of natural origin, men build terraces. The 
fields above the retaining walls may be large or the size of a 
table. They are hand-made fields. The dirt behind the walls 
washes down to lower levels and must be carried up again. The 
fields are ordinarily too small for pasturage and so manure must 
be carried up to them, usually on men's backs. Miss Semple has 
called this 'desperate agriculture.' 

The labor connected with terraces is not slight. Not only are 
walls built with stones brought together with great effort, but 
soil must be level where irrigation is practised. Irrigation is often 
necessary on terraces to maintain a reasonable water table. The 
soil, perhaps dug from below the surface zone, lacks humus and 
requires heavy manuring. Ordinarily the manure is brought up 
from the valley. Manure is heavy. 

The remarkable thing is the degree of perfection attained by 
primitive people as terrace engineers. Primitive negroes on the 
steep western face of the rift valley near Lake Tanganyika farm 
upon terraces. A savage negro tribe in the Murchison Range of 
Northern Nigeria have mapped their land in an elaborate terrace 
system. Some of the remarkable terraces of the world are in the 
mountains of Szechuen. Here in remote Western China whole 
mountain sides are devoted to terraces. In Western Tibet each 
lamasery has its terrace culture developed to high degree. The 
primitive peoples of Central Luzon are accomplished terrace 
engineers. No more surprising terraces are known than those of 
the Incas. At Salamanca in Peru the terraces extend up the 
mountains to the frost line. It has been estimated that in certain 
valleys the Incas had turned every possible usable slope into ter- 
races and that no room was left for an increase of cultivation or 
support of population. 

Terrace culture is not confined to primitive peoples. It is 
found generally in the mountains of Europe (Figure 49). The 
valley of the Rhone about Sierre, on the route of the Simplon 
Express, is carefully terraced. This section is mentioned because 



LAND UTILIZATION AND ECONOMICS 1 65 

it has been observed by so many travelers. Generally speaking, 
the drier the mountain slope the more important the terracing. 
The Eastern or Mediterranean Pyrenees have more terraces than 
the moister central portion. The Italian slope of the Alps has 
more terraces than the northern slope. Two reasons stand back 
of this. Agriculture on the drier slopes depends upon irrigation 
to a greater extent. Dry slopes have poor grass cover and are 
thus more subject to erosion. Grass is harvested in the gaves of the 
Central Pyrenees on slopes of 50 degrees without destructive soil 
erosion. (Figure 48.) 

The social implications of terrace culture and of irrigation are 
much the same as those arising from the communal ownership of 
pastures. There is, of necessity, a series of communal regulations 
controlling terrace building, replacing of soil, access to terraces, 
building of trunk canals, and regulation and distribution of 
water, and these force upon the people a group organization 
which tends towards the formation of syndicates and a socialistic 
attitude. 

The Factor of Relief: Andorra 

The writer has prepared a map of Andorra which illustrates 
strikingly the amount of territory which must be subtracted from 
the whole because of relief. The map shows areas which are so 
steep as to be barrens. Even the high peaks of this region are 
used as pasture where the slope angle is not prohibitive. Most of 
the steep areas are in a middle zone rather than the highest zone. 
(Figure 52.) There were no areal statistics in Andorra at the 
time (1928) this investigation was made. The map offered is the 
result of personal observation and of the study of the single topo- 
graphic map existing. This is on the scale of 1 150,000 with a 
25-meter contour interval. 

Four symbols are used on the map prepared by the writer. 
(1) The horizontal lines represent the area of continuous culti- 
vated fields on the valley bottoms or alluvial slopes, whether they 
are irrigated or farmed normally. Isolated fields, as fields upon 
the shoulders of the canyons, are not included. (2) The pointed 
symbol or inverted V indicates areas of accessible timber of com- 
mercial value. Over this area the stand varies greatly and for 
most sections the measure in board feet is not great. The tree 
which predominates and is of prime value is the mountain pine. 
The finer trees which once stood on the better watered areas in 



1 66 MOUNTAIN GEOGRAPHY 

the high mountains have been cut and the land given over to 
pasturage. (3) The vertical lines indicate cliff or land of over 100 
per cent (45 degree) slope. Much of this area has slight timber 
value; most of it is barren of even sheep forage, and some of it has 
talus which is a menace to the tilled lands below. (4) The dotted 
area is of upland pastures. The plas are grassy plateaus, of value 
chiefly for sheep. The comas are green parks devoted to cattle and 
horses. Many of the high points to the north are knobs rather 
than peaks, and here pasture — not of the best quality, it is true 
— covers the heights. Elsewhere it is topography, rather than 
climate, which limits the altitude of the grass. Land which is 
park land, that is, alternate forest and open area, has been 
mapped here as forest or pasture according as one or the other 
economy represents the predominant industry. 

A striking feature of the map is the amount of territory which 
must be considered barren because of too great declivity. This is 
a feature of land utilization which is commonly slighted in moun- 
tain mapping. A second noteworthy element of the map is that 
there are huge areas devoted to pasture. It is well that this should 
be emphasized, for certainly it stands first in the elements of the 
country's economy. The largest pasture area is to the north. 
This area is the little dissected portion of the summit plateau of 
the Pyrenees. Much of this area, though of high level, is flattish. 
Still a further factor in the large amount of pasture to the north 
is the altitude. The tree line is somewhere between 2000 and 
2400 meters. This latter contour encloses considerable areas of 
the north. 

The unequal distribution of the farmed area is a geological and 
topographical matter. Between the wider basins, the farmed 
land may narrow down to a mere strip of grass land lying in the 
deep shadowed gorge and chilled by the stream. Here precipi- 
tous walls prevent the farms from creeping up the valley sides. 
Where the valley is wider, the slopes are more gentle and the 
height limits of the fields are generally greater. 

The map may well stimulate interest in a more detailed de- 
scription of the landscape. The 495 square kilometers of the 
country are largely in one drainage basin; but the main stream, 
the Riu Valira, has two distinct branches and, in all, six open 
basins. These basins are caused by the unequal resistance of 
metamorphic rocks found in the southern part of the country. 
The northern portion, referred to previously as the pla country, 



LAND UTILIZATION AND ECONOMICS 167 

represents the granitic core of the range. As the valleys are 
gorges with slight bottom area, there is a great deal of the high 
level land represented in the southern portion. (Figure 51.) 
Here the uplands are of a different character, for glaciation has 
been much more severe in the more easily eroded rocks. Cirques, 
compound cirques, glacial lakes, and hanging valleys are com- 
mon. Both the northern and southern portions of the upland rep- 
resent summer pastures, but the organization of the industry is 
different in the two cases. On the plas large droves of sheep and 
goats, largely transhumants from France and Spain, feed in the 
summer time. The increased precipitation due to the altitude 
joins with the level surface and the impermeability of the granite 
rocks to provide excellent pasture for small stock. The highlands 
of the dissected southern portion consist of peaks and serrated 
ridges with steep cirque walls which separate cirque bottoms and 
high level valleys. Here the steep slopes have only enough soil 
and ground water for the mountain pine. The small areas of val- 
ley bottom soil offer excellent pasturage. On these comas the 
Andorran pastures his own beasts, largely cattle, mares, and 
young mules. These comas are not leased but are used by the 
communes. There are, then, two topographic levels of life in the 
country, the pastoral and the agricultural. The difference in re- 
lief between these two levels is 1000 to 1500 meters. In contrast 
to the mountain nomadism represented in many higher groups of 
mountains, as the Alps, ordinarily only a few shepherds take part 
in the seasonal movement. 

The valleys are distinctly glaciated. The forms of erosion are 
much more severe than in most parts of the Pyrenees, and the 
general aspect of the landscape is savage. The basin of Andorra 
la Vieja is a little Yosemite. Since glacial times there have been 
canyons cut which are still so youthful as to offer serious barriers 
to communication between the basins. Thus the gorge of Sant 
Antoine between the basin of Andorra and La Massana has to- 
day but a mule path hewn from the rocks. The gorge from An- 
dorra to Encamp has a road along its wall considerably above the 
stream (Figure 51), but from Encamp to Conillo the trail is 
forced up 320 meters over the shoulder of the mountain to pass 
the gorge of Meritxell. The climb is an arduous one for man or 
mule. The villages are on terraces which are lateral moraines. 
Whereas the glacial terraces in places offer high level agricul- 
tural land, elsewhere the morainic deposits are so rough and 



1 68 



MOUNTAIN GEOGRAPHY 



:#£ : 



'c*r.\ 



* RANSO \J\. • '^J^ZSt t Jfe 2778 
^iCaban 



^5a 




■«'»< 



hpIl^^ecsr'SB 


hJLJ \ Porteitle Blanche 
pniliiaiiP^ s. d Andorra 


PIA DC- 
UNQLA 


., / 



ANDORRA 

LAND UTILIZATION 

Uz=EI<vAwirff I — J Pasture Vj^I Forest mm slope 
Elevations in meters 



S KILOMETERS 
THE 6C06W BCVIEW. A*W 1^23 



Fig. 52. Land Utilization Map of Andorra 
The area of continuous cultivated fields on the valley bottoms or alluvial slopes is 
shown, whether irrigated or farmed normally. Isolated fields upon the shoulders of 
the canyons are not included. Under forest are shown areas of accessible timber of 
commercial value, chiefly the mountain pine; but the stand varies greatly and 
generally the board-foot measure is not high. 

Land of over 45 ° slope is distinguished: much of it has slight timber value; most is 
barren even of sheep forage; some of it has talus which menaces the tilled lands be- 
low. The upland pastures include the plas, grassy plateaus used largely for sheep, 
and the comas, green parks devoted to cattle and horses. Many of the high knobs of 
the north are covered with pasture of sorts to the summits; elsewhere it is topography 
rather than climate that limits the altitude of grass. Mixed forest and grassland has 
been mapped according to predominant use. The scale is approximately 1 : 290,000. 
(Courtesy of the Geographical Review, published by the American Geographical 
Society of New York.) 

strewn with huge boulders as to be unfit for use except for pastur- 
age between the rocks. On the whole, glaciation has improved 
the Andorran valleys as a place of habitation. The greatest con- 
tribution the ice made was in creating a terminal moraine on 
which Santa Coloma stands. A lake of short duration existed in 
the valley upstream. Alluvial deposits from the side walls have 
encroached upon the lacustrine soils. 



LAND UTILIZATION AND ECONOMICS 169 

The region is to be classed as of a moist Mediterranean climate 
in distinction to the dry Mediterranean climate to be found in 
Spain just without the borders of the country. Farming is both of 
the irrigated and the normal type. The mown grass, tobacco, 
and vegetables are grown by irrigation ; the cereals are a dry land 
crop. All level land or slopes that lie adjacent to and below 
canals on the valley walls are irrigated. These slopes are usually 
terraced and then are devoted to grass. As in all regions where 
beasts are pastured in the high mountains in summer, the great- 
est part of the cropped land is devoted to hay production for feed 
for the animals during the winter. Andorra has an excess of high 
pasture. Therefore the economic pressure for the hay land as 
against field crops is great. The more hay, the more beasts of 
Andorran ownership on the uplands. The number of cuttings of 
grass varies from one on the natural meadows at the edge of the 
plas to four on the irrigated fields. There is in Andorra, con- 
sidering the Mediterranean type of climate, a remarkable flow in 
the streams. That the uplands have a considerable duration of 
snow cover is an element in this condition, but the relative level- 
ness of the uplands as well as the number of lakes plays a part. 
There is always an excess of water for the amount of land to 
which it can be easily conveyed. 



BIBLIOGRAPHICAL NOTES 

Regional Works on Mountain Geography 

Allen, W. E. D. "The March-Lands of Georgia," in Geographical Jour- 

ml, lxxiv (1929), pp. ISS-^ 6 - 
Allix, Andre. "L'economie rurale en Oisans," in Revue de geographie 

alpine, xvii (1929), pp. 469-610. Stubborn agriculture. 
Allix, Andr6. VOisans. Paris, 1929. The thoroughness of the work is 

indicated by the fact that 800 pages are devoted to one small area of 

the French Alps. 
Atwood, W. W. " Utilization of the Rugged San Juans" (Colorado), in 

Economic Geography, iii (1927), pp. 193-209. 
Bernhard, Hans. Die Wirtschajtsprobleme des Vallemaggia (Tessin) als 

typischen Gebirgsentvolkerungsgebiet. Zurich, 1928. An excellent local 

study. 
Biermann, Charles. " La vallee de Conches en Valais," in Bulletin de la 

Societe Vaudoise des sciences naturelles, xliii (1907), pp. 39-175. Excel- 
lent. 



170 MOUNTAIN GEOGRAPHY 

Blache, Jules. Les massifs de la Grande-Chartreuse et du Vercors. Grenoble, 
1 93 1 . 2 vols. Blache is one of the promoters of the Grenoble school 
of Alpine geography, which, under the leadership of Blanchard, has 
caused the French Alps to be one of the best studied territories in the 
world. This work is a model for a regional study. 

Blanchard, Raoul. "Aiguilles," in Revue de geographie alpine, x (1922), 
pp. 127-165. 

Blanchard, Raoul. Les Alpes Jrancaise. Paris, 1925. Preliminary to the 
volumes upon the French Alps which are now being written by this 
master of mountain geography. 

Blanchard, Raoul, and Seive, F. Les Alpes franqaises a vol dfoiseau. Gre- 
noble, 1928. {Les beaux pays.) Illustrated by 137 heliogravures of 
mountain scenery. 

Bleicher, Gustave. Les Vosges, le sol et les habitants. Paris, 1890. 

Bouchet, Jeanne. "La vallee de la Roizonne en Taillefer: Enquete 
economique," in Revue de geographie alpine, xvi (1928), pp. 179-192. 

Bowman, Isaiah. The Andes of Southern Peru. New York, 1 916. A study 
in which carefulness of detail is combined with breadth of view. 

Brunies, Stephan. Der schweizerische Nationalpark, 3. Aufl. Basel, 1920. 
In the nature of a guide book. There is a French translation. 

Cavailles, Henri. La vie pastorale et agricole dans les Pyrenees. Paris, 1 93 1 . 

Cholley, Andre. Les Prealpes de Savoie et leur avant-pays. Paris, 1925. 

Dainelli, Giotto. Le Condizione delle Genti. Bologna, 1924. (Spedizione 
Italiana de Filippi nelP Himalaia, Caracorum, e Turchestan 
Cinese, 1913-14, Relazioni Scientifiche, serie ii, 8.) Anthropogeo- 
graphic material on Himalaya of a most scientific nature. 

Dainelli, Giotto. // Monte Bianco. Turin, 1926. Though it is not Dai- 
nelli's most serious work, nevertheless this is an informative book 
upon the Italian side of Mont Blanc. 

Desbuissons, Leon. La vallee de Binn (Valais). Lausanne, 1909. 

Flahault, Charles. La mise en valeur des terres pauvres par le boisement. 
Montpellier, 1920. 

Forrer, Niklaus. Z ur Anthr op geographie des Alpinen Thurtales. Bulach, 

1925- 

Friih, Jacob. Geographie der Schweiz- St. Gallcn, 1929-33. 2 vols. The 
great book on the geography of Switzerland. 

Gex, Francois. Dans les Alpes Jranqaises. Paris, 1929. 

Gibert, Andre. "Le Valgaudemar," in Revue de geographie alpine, xi 
(1923), pp. 663-782. 

Gos, Francois. Zjrmatt and its Valley, tr. by F. F. Roget. London, 1927. 

Gradmann, Robert. Suddeutschland. Stuttgart, 1931 . Chapter on the 
Bavarian Alps. 

Gsteu, Hermann. Beitrage zur Anthr op ogeographie von Vorarlberg. Stutt- 
gart, 1932. 

Hedin, Sven. Southern Tibet. Stockholm, 1916-22. 9 vols. Many other 



LAND UTILIZATION AND ECONOMICS I 71 

books of exploration could be mentioned. This is included as a 

splendid example. 
Hettner, Alfred. Die Kordillere von Bogota. Gotha, 1892. (Petermanns 

Mitteilungen, Erganzungsheft Nr. 104.) 
Krebs, Norbert. "Die bewohnten und unbewohnten Areale der Ost- 

Alpen," in Geographische ^eitschrift, xviii (1912), pp. 443-454. 
Krebs, Norbert. "Die Dachsteingruppe," in £eitschrift des Deutschen und 

Osterreichischen Alpenvereins, xlvi (191 5), pp. 1-42. 
Krebs, Norbert. Die Ostalpen und das heutige Osterreich. Stuttgart, 1928. 

2 vols. The great book on the regional economy of a mountain 

area. Instructive and philosophic. 
Krebs, Norbert. "Die Verteilung der Kulturen und die Volksdichte in 

den osterreichischen Alpen," in Mitteilungen der Geographischen 

Gesellschaft in Wien, lv (191 2), pp. 243-303. 
Leemann, Walter. " Zur Landwirtschaftskunde des Tavetsch," in 

Mitteilungen der Geographisch-Ethnographischen Gesellschaft in ^urich y 

xxix (1928-29), pp. 13-122. 
Levainville, J. "La vallee de Barcelonnette," in Annales de geographie, 

xvi (1907), pp. 223-244. 
Machatschek, Fritz. Die Alpen, 3. Aufl. Leipzig, 1929. A primer. 
Machatschek, Fritz. Landeskunde der Sudeten- und Westkarpathenldnder \ 

Stuttgart, 1927. 
Martonne, Emmanuel dc. Les Alpes. Paris, 1926. A review. 
Martonne, Emmanuel de. "The Carpathians: Physiographic Fea- 
tures Controlling Human Geography," in Geographical Review , iii 

(1917), pp. 4I7-437- 
Martonne, Emmanuel de. La Valachie. Paris, 1902. Deals with the 

Carpathians. 
Newbigin, Marion Isabel. Southern Europe. London, 1932. 
Ogilvie, A. G. Geography of the Central Andes. New York, 1922. (Ameri- 
can Geographical Society, Map of Hispanic America, Publication no. 

Reynier, Llie. "La region Privadoise," in Recueil des travaux de Pin- 

stitut de geographie alpine, iii (191 5), pp. 1-56. 
Robequain, Charles. " Le Trieves," in Revue de geographie alpine, x 

(1922), pp. 5-126. 
Roletto, G. B. "La Valle delP Orsigna," in Rivista Geografica Italiana, 

xxiii (1916), pp. 432-440; xxiv (1917). PP- 24~3 8 - 
Sieger, Robert. Die Alpen. Leipzig, 1900. 
Zur Geographie der deutschen Alpen Prof. Dr. Robert Sieger zur 60. Geburts- 

tage gewidmet. Vienna, 1924. Some of the essays in this Fest band are 

noted elsewhere. 
Sorre, Maximilien. Les Pyrenees, Paris, 1922. 
Sorre, Maximilien. Les Pyrenees mediterraneennes. Paris, 191 3. The more 

elaborate work. 



172 MOUNTAIN GEOGRAPHY 

Telker, Minnie J. "The Cascade Mountains," in Journal of Geography, 

xiii (1914-15), pp. 242-250. 
T6not, Suzanne. " Le massif de Belledonne," in Recueil des travaux de 

VInstitut de geographic alpine, vii (191 9), pp. 601-689. 
Tivollier, Jean. " Monographic de Ceillac (Hautes Alpes)," in Bulletin 

de la Societe d* etudes des Hautes- Alpes, xlv (1926), pp. 71-140. 
Vallot, Charles; Vallot, Joseph. Le massif du Mont-Blanc. Versailles, 

etc., 1921-23. 2 vols. An example of a scientific guidebook. 
Vila, Pau, and Chevalier, Marcel. La Cerdanya. Barcelona, 1926. See 

also the review by Philippe Arbos in Revue de geographic alpine, xv 

(1927), pp. 363~3 6 7- 
Wissman, Hermann von. "Das Mitter Ennstal," in Forschungen zur 

deutschen Landes- und Volkskunde, xxv (1927), pp. 1-144. 

Economic and Industrial Changes in Mountain Economy 

See also the works named in the preceding note. The French have a 
special fondness for historical studies of this sort. 

Allix, Andre. VOisans au Moyen-Age. Paris, 1929. 

Blache, Jules. " L'essartage, ancienne pratique culturale dans les Alpes 

dauphinoises," in Revue de geographic alpine, xi (1923), pp. 553-575. 
Blache, Jules. V homme et la montagne . Paris, 1933. 
Blanchard, Raoul. Grenoble: Etude de geographic urbaine. Paris, 191 1. 

The story of urban development in mountains. 
Blanchard, Raoul. "L'industrie de la houille blanche dans les Alpes 

franchises," in Annates de geographic, xxvi (191 7), pp. 15-41. 
Blanchard, Raoul. "The Utilization of Water Power in France," in 

Harvard Business Review, VI (1927-28), pp. 176-187. 
Brutails, J. A. Etude sur la condition des populations rurales du Roussillon au 

Moyen-Age. Paris, 189 1. 
Brutails, J. A. Notes sur V economic rurale du Roussillon a la fin de Vancicn 

regime. Perpignan, 1889. 
Clouzot, E. "A propos de la haute vallee du Ven£on a la fin du XV 8 

siecle," in Recueil des travaux de VInstitut de geographie alpine, vi (191 8), 

PP- 333-340. 

Coulter, J. W. "Land Utilization in the Santa Lucia Region," in Geo- 
graphical Review, xx (1930), pp. 469-479. 

Jorre, Georges. " L'am6nagement hydro61ectrique de la vallee d'Ossau," 
in Revue geographique des Pyrenees et du Sud-Ouest, ii (1931), pp. 317- 

336. 

Lorin, Henri. "L'industrie rurale en pays basque," in Le musee social, 
1906, Memoires et documents, pp. 349-375. The part of little in- 
dustries. 

Mathieu, Andr6. "Les petites industries de la montagne dans le Jura 



LAND UTILIZATION AND ECONOMICS 173 

frangais," in Annates de geographic, xxxviii (1 929), pp. 439-459. More 
of little industries. 

Mejean, Paul. "Le bassin de Bonneville, Haute-Savoie," in Revue de 
geographie alpine, xvi (1928), pp. 5-168. A valley given over to work- 
shops. 

Plande, R. " L'utilisation industrielle de la vallee d'Aspe (Pyr6n6es 
Occidentals), " in Revue de geographie alpine, xvii (1929), pp. 41—54. 
A development due to water power. 

Sclafert, Th6rese. Le Haut-Dauphine au Moyen Age. Paris, 1926. 



CHAPTER VIII 

MOUNTAIN POPULATIONS AND THEIR 
DISTRIBUTION 

General Aspects 

RELATIVELY speaking, mountains are agriculturally un- 
productive. Miss Semple has it that man, as a part of the 
mobile envelope of the world, like air and water, always feels the 
pull of gravity. Mountains, like the sterile portions of plains, 
have, for the most part, been occupied only after the more fa- 
vored lands. Some peoples, indeed, entered mountain regions at 
an early date, usually remnants of defeated folk who could not 
successfully resist their enemies in the lowlands. The Caucasus 
Mountains have so many such relics as to be called the graveyard of 
nations. Except for a tendency of the impoverished inhabitants 
to raid the richer lowlands, mountains have played a passive role 
in history. Once a race is established in a mountain fastness, iso- 
lation and defensibility of position have tended to preserve ethnic 
and cultural characteristics. Vidal de la Blache has said that 
mountains not only bring populations into being, but preserve 
them. The first part of the statement would seem hardly true, for 
mountains are, in fact, populated by the overflow of plains. The 
latter part of the statement is not true since 1850. Before that 
date the mountaineer was held to his home by the difficulty of 
transportation. The mountain spurs confined him to his valley 
settlement. But since 1850 the reverse has been the fact. 

Until 1850 mountain populations grew by natural increase. 
Population had become so dense that an economic crisis existed. 
One may read of economic conditions in a high isolated valley in 
the nineteenth century in the historical study by Allix of the 
Oisans, in the French Alps. Always in mountains tillable land is 
scarce and labor arduous. Crop returns are commonly pitiful. 
With populations increasing in certain valleys as greatly as 34 
per cent between 1800 and 1850, man power, to use an old phrase, 
was sorely exceeding bread power. This increase of population 
continued until the middle of the century, when a great flow of 
emigration commenced. The figures for Savoy are instructive. 



POPULATIONS AND DISTRIBUTION 175 





Population of Savoy 




Year 


Population 


Year 


Population 


J 7 2 3 
1783 
1828 


337.184 
423,166 
538,567 


1838 
I 9OI 
I906 
igil 


564,137 
518,514 

5 3, 02 7 



The figures in the above table representing the present century 
do not fully reveal the great decrease in rural population. Savoy 
includes some growing industrial centers. Gex calls attention to 
their growth. A purely rural region in Savoy is the valley of the 
Ubaye. Between 1851 and 1901 the population of this valley fell 
from 17,585 to 12,538 souls. Whole hamlets were abandoned. 
The same rural exodus is found in the Pyrenees. A few of the 
easily accessible valleys have increased in population during the 
last thirty years, because of the growth of industrial, commercial, 
or religious centers. But the higher and more remote valleys have 
decreased in population until some villages are only one-fourth 
their former size. Of the thirty-one mountainous departments of 
France, several increased between 1800 and 1850 at the rate of 
10 to 50 per cent. Between 1850 and 191 1 great losses were sus- 
tained. The Basses- Alpes lost 31 per cent, and Ariege, 26 per 
cent. In the 31 mountainous departments, 26, between 1901 and 
1906, lost 107,000 persons " because of soil erosion." Certainly 
soil erosion was a factor. But in this period the peripheral centers 
of Nice, Marseilles, Perpignan, Toulouse, Bordeaux, and Lyons 
all increased in size, no doubt at the expense of the adjacent 
mountain regions. Almost always a decrease of population in a 
mountain region denotes an increase of an urban center. This is 
usually a valley town. In the Val Pellice of the Italian Alps a 
population of 6000 in 1805 had increased to 9000 by 191 1 . In this 
valley is the town of Torre. The growth of the population of that 
center from 2000 in 1 805 to 5500 in 1 91 1 not only accounts for the 
increase, but indicates a slight decrease in the rural population. 

Improved facilities of transportation permitted and en- 
couraged the flight. Many authors concur in this opinion. This 
was the period of road-building and rail-laying. One is sur- 
prised at the number of communities at the beginning of this 
period without roads to the outside world. It has been estimated 
that Spain had 2000 such communities, largely in the mountains. 
Arbos gives a fragmentary table indicating the meager communi- 



176 MOUNTAIN GEOGRAPHY 

cation of certain regions of the French Alps and showing the pro- 
gress of road building. 

Progress in Road Building 

Total number Number of communes 
Region of communes without roads in 

i860 1880 1900 1914 

Canton of Chatelard (Bauges) 14 10 2 o 

Arrondissement of Albertville (Tarentaise, 

Beaufortin, Combe d'Arly) 42 25 20 1 

Arrondissement of Moutiers (Tarentaise) .56 42 36 4 2 

Arrondissement of St. -Jean-de-Maurienne 68 45 33 18 11 

Arrondissement of Castellane 48 18 8 4 

Canton of La Javie (Pr6alpes de Digne) ..10 633 

It needed but a few roads to engender more. Road-building 
went forwards by leaps and bounds in the whole of the Alps. 
After the advantages of better communication had been demon- 
strated, the passive mental attitude of the mountaineer quickly 
changed. Ease of communication in mountain regions is not 
merely a function of topography and relief, it is an expression of 
the economical and mental needs of the mountaineer. 

It is possible that the nomadic life of mountaineers, especially 
of the Swiss, prepared them psychologically for the emigration. 
In addition certain valleys, because of meager economy, had long 
practised the custom of sending forth mercenary soldiers, ped- 
dlers, itinerant merchants, and theatrical troupes. One writer, 
Miss Semple, in referring to these migratory habits, rather over- 
states the matter in saying that mountains, like the sea, breed 
travelers. These annual migrants were the exception rather than 
the rule. With the coming of the modern period and better trans- 
portation, there was indeed a factor which tended to keep people 
at home. This was the increasing importance of the cottage in- 
dustries. An element outside of the mountains directly respon- 
sible for a large part of the emigration was the attraction of new 
lands in the Americas. Mountain people, living under as great 
economic pressure as any in Europe, logically took part in the 
American settlement. Sections of the Haut-Diois in the French 
Alps, which lost in two decades as much as 30, 40, and 44 per 
cent of their populations, sent great numbers to Mexico and the 
United States. Latin America took its quota of mountaineers. 
Another factor in the emigration was the general flight from the 
land towards the cities, a migration common to all European 
rural districts at this time. This, we have pointed out, did not 



POPULATIONS AND DISTRIBUTION 177 

greatly decrease the populations of districts where cities existed 
on valley plains within their limits. 

Maull has said, "The upper limit of population distribution is 
the battle ground of the denser centers of population." 

Population and Altitude 

If an average were taken for all European mountains, it would 
be fair to presume that increasing altitude would show decreasing 
population. Five per cent of the population of Switzerland is 
found above 1 000 meters, and yet the area lying above that eleva- 
tion is a considerably larger proportion of the total area of the 
country. Within limits, certainly below the climatic snow line or 
zones of long snow duration, the scant population of the upper 
levels is a result of the meager resources. The relation between 
meager resources and paucity of population is even more direct 
than that between percentage of level land and density of popu- 
lation. The severity of the decrease with ascent is well illustrated 
by the canton of Grisons, which has a density of population only 
a thirty-fifth that of the canton of Zurich. Most of Zurich is on 
the Swiss plain, while Grisons is mountainous upland. The up- 
land is incapable of supporting large numbers of people. The 
climate is rude; harvests are limited in quantity and character; 
much of the area is actually barren. In the high cantons of 
Uri and Valais, Switzerland, more than half of the area is 
barren, and this is true of the Austrian provinces of Tirol and 
Carinthia. 

The decrease of population with altitude is not, however, a 
regular matter (Figure 61). A reason for this is that zones of in- 
creasing altitude do not always show a decreasing acreage avail- 
able for habitation. Forest zones and steep zones are more 
thinly populated than open lands and gentler slopes at higher 
altitudes. There are exceptions. Most zonings of population are 
taken in a valley cross section. If one took the density of popula- 
tion of valley floors alone, and arranged the densities according 
to increasing altitudes along the valley, a much greater regularity 
of decrease would appear. 

Blanchard has one of the more exact studies of altitude and 
population. He has figured the mean altitudes of certain zones in 
the French Alps and the percentage of those zones which are in- 
habited. There is an approximate regularity of decrease only up 



1 7 8 



MOUNTAIN GEOGRAPHY 




POPULATIONS AND DISTRIBUTION 179 

to 1 900 meters, and even that must be looked upon as a chance 
occurrence. 

Inhabitation by Altitude and Mass 

Mean altitude Percentage of inhabitation 

1 408 58 

I419 56 

1433 58 

I 65O 40 

1660 35 

1 7 1 2 40 

1722 40 

1834 23 

igi I 56 

1954 50 

I980 27 

2I05 42 

2160 37 

2203 34 

2429 26 

2556 12 

2396 15 

2279 31 

Populations in tropical mountains are again another matter. 
Lowlands there have exhausting and debilitating climates be- 
cause of heat and excessive humidity. Elevations of 1000 meters 
and above are more favorable. India is said to be ruled from 
over 2100 meters. The majority of the people of Ecuador and 
Bolivia live above the contour line of 2000 meters. In Peru there 
is permanent habitation between 3200 and 4500 meters. One 
shepherd's home is reported by Bowman at 52 1 5 meters. Where- 
as not one-fifth of the population of Grisons live above 1500 
meters, the mountain mass of Ethiopia, in equatorial Africa, 
recently supported five chief cities between the elevations of 1857 
and 2500 meters. 

South America illustrates the fact that in inter-tropical regions 
elevation makes for health and greater human activity. 

Heights of Settlement in Peru 

Cuzco (15,000 inhabitants) 3200 meters 

Sicuani ('the Paradise of Peru 5 ) 3534 " 

Cruciro 3953 " 

Indeed, in Peru the most densely inhabited zone is between 
1500 and 3500 meters. There is Cerro de Pasco, a min- 



i8o 

NW 



w 



MOUNTAIN GEOGRAPHY 
5W S SE E NE 



*s 








Z\ 


OO 


"""""'"v 
















\ 




S 








19 


00 






























17 


00 












^ — '•*• 

/ 

/ 


-^ 












/ 
/ 
/ 


/ 

/ 
/ 


ift 


00 ••. \ 










/ 
/ 
/ 
/ 
/ 






\ 
\ 

\ \ 

'. \ 
■•. \ 
•■ \ 
\ v 








••' / 


/ 

/ 
/ 

/ 




13 


00 


\ 
\ 






; / 
./ 










\ 
\ 

\ \ 

N 


--' 


^ -. 


•*' 






MOO 


m tries 






* 



Houses- 

Gram: 

Forests: 



Fio. 54. 



Permanent House Limits and Grain and Forest Limits in the 
Ortler Alps, Austria. (After Fritsch.) 



ing camp, which has 13,000 souls living at 4350 meters, far be- 
yond the timber line. Bolivia is largely a land of a high plateau. 
The climate is far more healthful than that of the Amazonian 
lowlands. The following cities have surprising size for their 
elevation: 

Heights of Settlement in Bolivia 

City Population Altitude 

La Paz 100,000 3600 meters 

Oruro 22,000 361 7 " 

Potosi 29,000 4000 " 



POPULATIONS AND DISTRIBUTION 181 

The altitude of human occupation in Tibet is, in view of the 
principle of economic returns diminishing with elevation, almost 
a paradox. Villages dependent upon tillage are common at a 
height of 4000 meters. One finds clusters of permanent dwellings 
up to 4880 meters, where not only herdsmen but agriculturists 
are found. Barley is grown at 4400 meters, vegetables at 3600 
meters. Crops at these extreme elevations are in response to low 
latitude, the dryness of the air, high insolation, and irrigation. 

Some high settlements in the Alps are worth note: 

St. Moritz 1856 meters 

Arosa 1892 " 

Chanolin 1 936 " 

Cresta 1 956 " 

Findelen 2075 " 

Juf 2133 

St. Moritz and Arosa, both in Grisons, originally villages depend- 
ing on the land, are now great health resorts. Chandolin in the 
Val d'Anniviers is, curiously enough, above many of the pas- 
tures, so that the cattle in spring begin their migrations by de- 
scending. Findelen, in the same valley, has an anomalous agri- 
culture. There, on steep slopes facing the sun, rye and barley 
ripen. Immediately opposite, on a shady slope, is a belt of Arctic- 
Alpine plants, and near by are two glaciers. Cresta has but five 
houses, while, at the end of the century, neighboring Juf, holding 
the Swiss record for altitude, had but 24 inhabitants. One sum- 
mer village within the Val d'Anniviers, Alp de Lona, lies at the 
height of 2665 meters. 

Details of Population 

Among the factors which control the details of distribution of 
population in mountains are altitude, relief, mass of mountain, 
distribution of critical temperatures, distribution of rain, dura- 
tion of snow cover, winds, nature of bed rock, thickness and char- 
acter of the soil, character of the vegetation, danger of floods, 
degree of accessibility, height of mountain passes, economic con- 
ditions, and the social character of the people. The question 
comes early to mind as to how high man may go in mountains 
before he meets physiological difficulties. In Bolivia there is a 
colony of miners living at 5000 meters. Here the barometric 
pressure approximates half of that at sea level. In Tibet there are 



1 82 MOUNTAIN GEOGRAPHY 

permanent hamlets and agriculture of a sort at 4400 meters. 
Matters such as heart palpitation and mountain sickness seem of 
minor importance if men mount slowly, by generations as it 
were. Adjustment to very high altitudes, certainly of selected 
persons, seems quite possible. In European mountains, however, 
the mountain top or the permanent snow line is reached before 
physiological limitations are met. We shall discuss health and 
altitude in a later portion of the chapter. 

In Alpine Europe, groups living at high altitudes are agri- 
cultural or pastoral. The so-called permanent or agricultural 
dwellings of these people are not often at great heights. The 
traveler may easily obtain false impressions as to the altitudinal 
distribution of the people and the density of population if the 
temporary or pastoral shelters are not eliminated from considera- 
tion. The number of buildings is not evidence of the distribution 
or numbers of inhabitants. Many peoples have a complete per- 
manent village on the valley floor and another at the lower edge 
of the pastures. One Swiss group actually has four villages. At 
Courmayer, on the Italian side of Mont Blanc, each family has 
a dwelling and a barn in the valley village, a summer chalet 
higher up the valley, and on the high alp a herdsman's hut. We 
cannot, therefore, take the number of houses as indicative of the 
size of population, nor can we take the highest houses as indica- 
tive of the economic limits of 'permanent* habitation. Though 
numbers of people go to the higher villages for short periods some 
two or three times a year, we must regard the permanent resi- 
dence as that in the lower valley. Also the number of people who 
follow the herds to the higher pastures is small. One of the few 
noteworthy exceptions is in portions of the Carpathians, where 
the summer population is sixteen to the square kilometer. The 
houses at the lower levels are the primary homes, but their inhab- 
itants have an extensive zone of activity. This is a broad zone on 
the mountain flanks, most of which is above the dwelling. The 
high-level temporary villages are auxiliary homes. In Switzer- 
land the permanent house limit is perhaps 1 00 meters below the 
forest line, and 800-900 meters below the permanent snow line. 

In the Alps the details of distribution of population are so 
many and so interwoven that it is difficult to state the situation in 
a paragraph. The complexity is suggested by a French study of 
the Italian Val d'Aoste. This valley, with an east-west trend, has 
a sunny and a shady slope, and great geologic and topographic 



POPULATIONS AND DISTRIBUTION 183 

variety. Here are ancient crystallines, glacial deposits, terraces, 
and alluvial cones. The valley floor has the greatest amount of 
level land and the greatest number of inhabitants. On 7 per cent 
of the land one finds 18 per cent of the people. The second zone 
of concentration is one common to many Alpine valleys. This is 
the zone of the slope villages. Here on 3 per cent of the area live 
9.5 per cent of the people. The third zone is the region of a com- 
mercial town, for the valley is on a route into France. This dis- 
tribution is typical for east- west valleys; a dense concentration in 
the valley bottom, and a secondary concentration midway on the 
sunny slope. The shady slope is ordinarily little inhabited. 
Topographic and geologic conditions complicate matters, as do 
a number of climatic factors. There cannot be a simple state- 
ment of details of distribution of population in Alpine mountains. 

A definite limitation to the density of rural population in 
mountains and a factor in the distribution is the restriction of 
productive land. Sixty-six per cent only of Alpine Europe is pro- 
ductive. In the Grisons, Switzerland, but 40.6 per cent is produc- 
tive. Much of this productive area is, in the nature of things, 
pastoral land and supports relatively few people. The Alps as a 
whole support some twenty-five persons to the square kilometer, 
but if we consider only the productive areas, the population is 
forty to fifty persons to the square kilometer. The density of 
population on the valley bottoms is high as compared even with 
agricultural regions of the plains. But relief and barren ground 
separate these settlements. Mountain populations are generally 
in isolated chambers or corridors. 

A study has been given us by Tonilio of the upper Valcamo- 
nica, which, because of its detail and because of the exceptions 
noted, is especially worth our attention. The valley is an east- 
west valley and is, a most important fact, a route to the Passo del 
Tonale. Within the fifteen kilometers which comprise the study, 
the elevation of the valley floor mounts from 850 meters to 1883 
meters. The summits of the flanking walls mount to 3554 and 
3278 meters. Geologically, the region is in the schist which 
flanks the crystalline massif. The valley walls show evidence of 
three terraces, the highest of which is pre-glacial. The elevations 
of the terraces are 1 150-1376 meters, 1450— 1800 meters, and 
1600-2000 meters. The valley floor is divided into three basins. 

The valley is overpopulated. Considering the total area of the 
drainage basin, the population is but 29 persons to the square 



1 84 MOUNTAIN GEOGRAPHY 

kilometer. But in terms of the land cultivated the density is of 
394 to the square kilometer. In spite of this large population 
there was a twenty-five per cent increase in the valley from 1 86 1 
to 191 1. This period was one generally of depopulation in 
French Alpine valleys. Previous to the commencement of the 
World War in 1914, the surplus population was supported by 
seasonal migrations to Germany and Austria in search of work. 
For this purpose each summer a third of the men left the valley. 

Practically all of the people live on land of little slope. Thirty- 
nine per cent of the people live on alluvial fans, 38 per cent on 
fluvial glacial terraces, 16 on orographic terraces, and but 
7 per cent on mountain slopes. 

As we have said elsewhere, most people in high valleys ' live in 
the sun, J that is, on the adret exposure. The table on page 185 
is from Mademoiselle Vessereau, and needs no discussion. 



Economic Factors 

It might be supposed that the building of roads and the laying 
of rails into mountain regions after 1850 would improve the 
economic conditions there and so encourage the people to re- 
main. The outward flood was not, however, to be easily stopped. 
An economic revolution within the mountains did take place, 
but it was the direct result of decreased population rather than of 
improved communications. Hitherto agriculture had been a 
mixed farming of the subsistence type. With the decrease in farm 
hands it necessarily changed to a pastoral and dairying economy. 
Grain areas decreased. The high fields, laborious to till, were 
abandoned first. In the Metnitztal, Austria, the highest grain 
field fifty years ago was above 1400 meters. Today there is none 
in the valley above 11 50 meters. True, railways in mountains 
have made possible the use of chemical fertilizers and so doubled 
the production per hectare, but the decline of population is still 
evident. Generally the more precarious the regional economy, 
the greater the recent decrease in population. This is well illus- 
trated by Tenot in a study of the Belledonne Massif. Another 
statement of rather general application is that the more remote 
the commune, the greater its decrease of population. By the 
same token, satellite communities suffer depopulation more 
severely than the commercial centers. 

The economic crisis due to overpopulation had literally left 



POPULATIONS AND DISTRIBUTION 



185 



Proportion of Population Living on the Adret and on the 
Ubac in Certain Valleys 

(Name* of north-south valleys are underlined. In these cases the western exposure is reckoned the adret.) 



Pop. on 

Valleys studied adret 

Aigue Blanche (Queyras) 536 

Aigue Agnelle u 272 

Cristillan u 344 

Tarentaise, from Tours to l'Arly ... 513 
Tarentaise, from Bourg-St. Maurice 

to Ste.-Foy 191 9 

Torrent d'Arvieux in Queyras .... 495 

Oetztal ? 

Torrent de Landry in Tarentaise . . . 950 

Pongau 843 1 

Pinzgau 9645 

Doron de Beaufort 2942 

Valde Conches 3124 

Guil 1966 

Pellice 7405 

Cluson 11 744 

Tarentaise, from Bourg-St. -Maurice 

to Moutiers 6570 

Vintschgau 1 1272 

DoireRipaire 61582 

Val d'Anniviers *498 

Valais (according to Lugeon) 26000 

Tarentaise, from Moutiers to Tours 4672 
Maurienne, from Montsapey to the 

confluence 2436 

Maurienne, from the source to St.- 

Jean 1 1 729 

Maurienne, from St. Jean to Mont- 
sapey 7886 

Val d'HeYens 3 1 41 

Tarentaise, from Val de Tignes to 

Ste.-Foy 920 

Inntal 33763 

Vallouise 1 1 48 

Val de Suse 26072 

Val d'Ossola 5618 

Val Chiese 3020 

Tarentaise, from the source to the 

Val de Tignes o 



Pop. on 

ubac 


Total pop- 
ulation 


Propor- 
tion of 
the pop. 
• on the 
adret 




O 


536 


IOO ] 


per cent 


O 


272 


IOO 


u 


O 


344 


IOO 


u 


O 


5 l 3 


IOO 


u 


63 


1982 


96.8 


u 


40 


535 


92.5 


u 


? 


p 


91 


u 


1 IO 


1060 


89.6 


u 


I504 

2036 

642 

868 

585 
2360 

4355 


9935 
11681 

3584 
3992 
2551 
9765 
16099 


84.8 
82.7 
82 
78.2 

77 

75-8 

72.9 


u 
u 
u 
u 
u 
u 
u 


3074 

5317 

27084 

755 


9 6 44 
16589 
88666 

2253 


68.1 

67.9 
69.4 
66.4 


u 
a 
u 
a 


14000 
2791 


40000 
7463 


65 
62.6 


u 

a 


1472 


3908 


62.3 


u 


774i 


19470 


60.2 


u 


5 8 43 


J 3729 


57-4 


u 


2671 


5812 


54 


u 


832 


1752 


52.5 


li 


33517 


67280 


50.1 


« 


"57 
39264 
13810 


23°5 
65336 
19428 


49.8 

39-9 
28.9 


« 
u 


1 1272 


14292 


21. 1 


u 



271 



271 



1 86 



MOUNTAIN GEOGRAPHY 



scars on the mountain sides which were difficult to heal. One of 
the most serious of things that may happen to a mountain region 
is excessive deforestation. Trees had been ruthlessly cut for lum- 
ber, fuel, and for clearings for fields and pastures. Deforestation 
means excessive soil erosion of the slopes, torrential deposits on 




Fig. 55. Population about Dorfli, Switzerland 

Note how many houses lie on the adret. Scale 1/50,000. Contour interval 

30 meters. 

valley fields, and loss of ground water. Danger of landslide and 
avalanche is increased. Moreover, the tilled fields had been over- 
cropped and impoverished. The pastures had lost area to make 
way for crops. The cattle, confined to smaller areas, were under- 
weight. Manure for the fields was lacking because of the decrease 
in livestock. Let me give an extreme example of decay. On 
October 28, 1888, the inhabitants of Chaudun, in the French 



POPULATIONS AND DISTRIBUTION 187 

Alps, petitioned that the state acquire 2027 hectares of the moun- 
tainous portion of the commune, which was so sterile that the 
inhabitants were starving. The commune was purchased and 
politically abolished in 1895. 

The increased prosperity expected from the betterment of 
transportation did come, if tardily. Since 1 900 conditions in even 
the remote valleys are decidedly improved, in spite of the flight 
from the land characteristic of the post-war period. The Haut- 
Diois, previously referred to, is an example in case. Railways 
have made markets accessible and have brought in chemical 
fertilizers. Sown meadows are now increasing in area. The yield 
of cereals per hectare has increased. Animals are now raised for 
shipment and there is a local slaughter-house. Previous to 1 900, 
the region was poverty-stricken. Now there is an increasing 
prosperity and we may expect increasing population. 

The valley of Roizonne, in the French Alps, illustrates the 
economic struggle of high and isolated valleys. This valley is 
oriented north-south and so has no true adret. The soil is poor. 
The climate is rude. Snow lies on the ground from the end of 
November to the last of March. Bouchet tells us how the igno- 
rance bred of isolation has delayed until recently the use of arti- 
ficial fertilizers. Agriculture suffers from depopulation. The 
young people making trips in the winter to the factory towns 
beyond the mountains have been unable to return. Valnoire (a 
sinister name) sixty years ago had 30 people. Today it is aban- 
doned. La Morte (also sinister), though the highest commune in 
the valley and surrounded by desolation, is the best preserved 
because it has the best economy. Agriculture has decreased in 
importance, while pastoral pursuits have gained. In the last 
decade, however, the use of chemical fertilizers has been leading 
to a revival of cultivation. There is hope for the future. 

Location of Dwellings 

The location of dwellings is, of course, decisive in the distribu- 
tion of population. An excellent study on this question is that of 
Fliickiger. Valley by valley, rather than by massifs, he takes up 
the Swiss mountain areas and analyzes in a masterly fashion the 
factors in the distribution of population. With the Siegfried 
Topographical Atlas of Switzerland as a basis and with the aid of 
ImhoPs study on forest line and Jegerlehner's study on snow line, 



1 88 MOUNTAIN GEOGRAPHY 

he brings out the climatic, orographic, and economic factors. 
He sees two zones of population: one the zone of permanent 
houses, and the other the zone of temporary houses. It is the 
permanent house upon which the census is based, and in this por- 
tion of the chapter we shall not concern ourselves with the houses 
of the upper zone. 

The factors influencing forest line and snow line likewise play a 
part in fixing the upper limit of elevation of isolated permanent 
houses, but, because in the limit of houses there is the element of 
human choice, the parallelism of the several lines is not exact. 
Just as with the snow line and forest line, the greater the mass the 
higher the dwelling line. Generally house lines are ioo meters 
below the forest line and 800 meters below the orographic snow 
line. But man often makes a poor adjustment of house location 
to climatic conditions because of economic or social factors. He 
may, as sometimes along pass routes, locate his house in defiance 
of climatic conditions. On the other hand, the Italian dislikes 
isolation, and so, living in villages, fails to elevate his house site 
to the climatic-economic limits. 

Exposure is of utmost importance in the location of houses 
which stand apart. In mountains of middle or high latitudes, 
people having the choice always live in the sunlight. Many an 
Alpine valley has no settlement and few houses on the shady side. 
Even those villages that are on the valley floor have a smaller and 
less valuable portion of their land area on the ubac side of the val- 
ley stream. Houses of elevation on the ubac slope are ordinarily 
upon secondary adrets. This is true, for example, of the two 
slopes of the Brenner. Houses of surprising elevation are to be 
found in high valleys favorably situated for insolation. 

In the French Pyrenees of Ariege in two valleys there is the 
following distribution of dwellings: 

Per Cent of Dwellings in Two Valleys According to Exposure 

Region Adret Valley floor Ubac 

Sillon Ariegeois ... 42 46 12 

Vic de Sos 27 37 36 

Of the first situation, all of the dwellings on the ubac were on 
secondary adrets. In the second situation, the 36 per cent on the 
ubac included 26 per cent on secondary adrets. The secondary 
adret is used as the site for isolated houses much more frequently 
than for villages. 

Houses are located with due appreciation of inversion of tern- 



POPULATIONS AND DISTRIBUTION 189 

perature and the consequent thermal belts. The depth of snow 
in the valley bottom has some influence on locations. Cold winds 
from glaciers have caused to be left uninhabited upper portions of 
valleys which might otherwise hold a house site. In the Pinzgau, 
24 of the settlements are on fans, 9 in basin and valley bottoms, 
and 7 on terraces. 

The topographic factors are many. A glaciated valley may 
have all houses confined to the exact valley bottom, because the 
slope of the valley wall is everywhere too steep for houses. Where 
slopes are gentle, houses may mount to a considerable elevation 
above the valley floor. If there is little level land in a valley, the 
houses will be on slopes in order to leave all level land for fields. 
Extreme cases of this are the ' piled ' houses or Ladakh, which 
literally cling to cliffs rather than encroach upon tillable land. 

Not a little of what is said here is of necessity repeated later in 
this chapter. There are social factors which enter into the ques- 
tion of the isolated dwelling as over against the hamlet or village. 
It is well known that in uninterrupted plains there are regions of 
villages and regions of isolated farmsteads. Historical conditions 
and habits of gregariousness affect racial choice. Generally the 
Italians avoid the isolated dwelling to a greater extent than do the 
German mountaineers. The man who depends to an extent upon 
wood-cutting for a living will live apart. The scarcity of springs 
or water supply on certain slopes will force people to live in vil- 
lages where water is accessible. These conditioning factors apply 
particularly to the permanent habitations. Yet another set of 
conditions control the location of the alp homes and shelters. It 
is hoped that Chapter VII deals sufficiently with these. 

Towns and Villages 

The height limit of towns and villages is affected by some of the 
same conditioning factors as limits of dwelling sites or of culture 
and vegetation. Particularly do these limits have climatic fac- 
tors in common. There are some topographic factors which are 
peculiar to towns as distinct from hamlets or isolated dwellings. 

To recount completely the climatic factors in the location of 
towns would be to repeat much that has been said in previous 
sections. But once again the two elements of exposure to sun and 
mass of mountain should be noted for their almost dispropor- 
tionate importance. The greater the elevation the more impor- 
tant the matter of exposure. The high valleys of the Hohen 



I go MOUNTAIN GEOGRAPHY 

Tauren have 90 per cent of their village population living on the 
adret. Frequently the percentage of population living on an adret 
of a valley is not one hundred simply because the villages are on 
the valley bottom and some people live immediately across the 
stream. The settlement across the stream is technically on the 
ubac and may have a separate name. People living in the shade 
are frequently socially looked down upon. Mass of mountain has 
the same climatic significance and topographic implications in 
the location villages and towns as in limits of culture. There are, 
of course, exceptions. Meolans in the French Alps lies on a steep 
ubac. On the day of the winter solstice this town experiences but 
one hour of direct sunlight. Rochebrune, another ubac town, has 
but four and one-half hours of sunlight on the day of the winter 
solstice. 

Sieger gives the following table showing factors of exposure in 
the height limits of permanent settlements: 

Height Limits of Settlement and Exposure 

Adret Ubac 

Runtental I2 44 m - I2 33 m - 

Echddergraben .... 1236 1 172 

Ratchtal 1 140 1080 

Fliickiger says that on the average in all of Switzerland the 
adret settlements are 75 meters higher than the ubac. Lugeon 
offers figures to show that the greater number of settlements are 
in the sun. In the upper valley of the Rhone 34,000 people live 
in the sun and 20,000 are on the shady side of the river. The val- 
ley of Conches, a high valley in Valais, having an east-west axis, 
has flanks of almost equal inclination. Three thousand of the 
mountaineers of this valley reside on the adret, whereas but 700 to 
800 persons live on the ubac. 

The topographic factors deserve rather special attention. Vil- 
lages may not always require level land, but towns ordinarily do. 
Moreover, towns require considerable area as garden space. 
Hence towns are ordinarily situated in open basins. Inner- 
kirchen, in the Bernese Oberland, is located where a basin has 
been formed by two tributary streams joining a main stream. 
There is a verrou, or barrier of bed rock across the valley, imme- 
diately below the town. Luz in the French Pyrenees is in a basin 
which is typical of the valleys of that range. Arosa lies in an open 
basin at the head of a main valley and so has the advantage of 
long hours of sunlight. 



POPULATIONS AND DISTRIBUTION 191 

A most common location for towns lies on contact zones be- 
tween regions of contrasted culture. In the Sierra Nevada of 
Spain, practically all the towns lie along a geological contact, 
because here is the juxtaposition of two cultures founded upon 
the geological contrasts. Towns in mountains frequently lie along 
the zone between vine and chestnut culture, woods and pasture 
culture, vine and wheat culture, valley and slope culture. 

Several zones of town and village settlement within a single 
valley are not uncommon. The valley floor is likely to be the 
chief zone, though where there is a post-glacial gorge the con- 
centration there may be lacking. Above the valley bottom there 
may be terraces, structural or superficial in origin, which hold 
lines of settlement. Such terraces may be found on both sides of 
the valley or upon only one. A terrace may afford communica- 
tion along its course, and so bind those villages together as a 
series of communes. Yet communication with the valley floor 
may be of such difficulty as to develop separate cultures for the 
terrace village and the valley village lying plainly in view. 

The menace of catastrophes in mountains limits village sites. 
Areas of frequent avalanches are out of the question. Occasional 
avalanches are guarded against at no small cost to certain com- 
munes. Floods keep many villages to the cones and fans, and so 
above the flood mark. Thus Chur, Trimmis, and lesser towns lie 
upon a single cone of dejection which has been washed down 
upon the plain of the upper Rhine valley. The towns are well 
above danger of flood. Towns have been known to be under- 
mined by soil creep or land slide. Advancing talus on cones re- 
stricts the location of settlements at the foot of steep cliffs. Lastly, 
winds influence town location. Cold mountain winds that occur 
with the setting of the sun may make location near the stream 
unsuitable. Indeed the stream itself can be a refrigerating in- 
fluence. Elsewhere winds in gorges or in line with a pass may be 
too violent for comfort. Hamlets have had to be abandoned be- 
cause of strength of winds. 

Altitude and Health 

There is a physiological limitation to the altitude at which man 
may live with comfort. Potosf, in the Andes, situated at 4000 
meters, once had a population of 150,000 inhabitants. Cerro de 
Pasco in the same mountains, with 13,000 persons, lies at 4350 
meters. This is higher than the peaks of most European moun- 



192 MOUNTAIN GEOGRAPHY 

tains. But were these people healthy? Could they work effi- 
ciently? Could children be born in safety? Was permanent 
group acclimatization possible? The answer is that those who 
came to the high elevations without gradual adjustment were 
unable to work at full capacity without strain on heart and lungs. 
Those with weak hearts died. Pneumonia or lung troubles were 
likely to be fatal. At these high stations childbearing for women 
of the valleys was dangerous for the mother and usually fatal for 
the child. There is, on the other hand, no reason why any race 
may not live at high altitudes if it becomes acclimated gradually 
through the course of generations. Adjustments of the body will 
come about, and there will be a racial adjustment to the altitude 
by the processes of selection. 

Rapid ascent of mountains results in an illness similar to that 
experienced by aviators and men subjected to artificial reduction 
of pressure. The fact of mountain sickness (altitude illness, puna, 
soroche, chuno, mal de montagne, and Bergkrankheit) has been long 
known, but the scientific study of its various aspects is rather re- 
cent. Medical authorities, mountaineers, and others still differ 
greatly concerning the causes and effects of mountain sickness, 
the usual height at which illness first appears, the height at which 
it is most prevalent, and the possibility of man's acclimatization 
to high altitudes. The symptoms of mountain sickness are dizzi- 
ness, nausea, accelerated respiration and heart action, dimness 
of vision, and inability to move. Mountain sickness presents a 
field of study as complicated as those of mountain climates, life, 
and vegetation. Conditions vary so greatly, and the factors in- 
volved are so closely interwoven, that it is very difficult to isolate 
the causes of symptoms in any particular case. Symptoms vary 
greatly in different persons and under different conditions. 

Some assert that the deficiency of oxygen at great heights is the 
explanation of all symptoms. The result is an anaemic condition 
in the blood. A high haemoglobin content, and thus an increased 
capacity to absorb oxygen, is characteristic of both animals and 
humans living at high altitudes. Visitors, as they become accli- 
mated, may have an increase of red blood corpuscles of more than 
50 per cent. 

That persons arriving at altitudes without exertion, as by 
train, are often affected with illness, somewhat contradicts the 
theory that mountain illness is the result of fatigue. Nevertheless 
there is support for those who believe that fatigue is a factor in 



POPULATIONS AND DISTRIBUTION 193 

mountain sickness. Severe exertion does aggravate such illness, 
particularly in the matter of heart action. The writer has often 
worked without illness at elevations considerably over 3350 me- 
ters, and yet at that elevation he once suffered acute nausea. 
This was on an ascent of Mount Etna. The climb had been 
largely in sliding volcanic ash. The fatigue of the writer and an- 
other man had been unusually great in the climb on foot from a 
station at 600 meters. A woman companion had ridden the worst 
of the climb on muleback. She was unaffected, whereas the men 
were seriously affected by nausea and palpitation of the heart. 
A third theory looks to diminished pressure as the main cause of 
the illness. From these several viewpoints a great variety of ex- 
planations have been evolved. Each theory seems to account for 
some of the symptoms but not for all. The statement that reduc- 
tion both of oxygen and of pressure plays an important part in 
causing mountain sickness is probably nearest the truth. 

A phase of the theory of pressure is that the sickness is caused 
by the effect on the organs of the body of the loss of equilibrium 
between the pressure of the external air and the gas of the in- 
terior of the body. Among other things, it is thought that di- 
minished pressure disturbs the circulation in that portion of the 
spinal cord immediately concerned with the nutrition and muscu- 
lar control of the legs. The partial loss of power of the limbs is 
due to the collection and stagnation of blood !n these portions. 
Perhaps the change of pressure produces a more contracted state 
of the lungs, causing a stagnation of the venous circulation, the 
resulting shortage of oxygen in the body being the cause of the 
symptoms. The extreme of pressure in its relation to man is 
found at San Vincente in Bolivia at 4575 meters, where the pres- 
sure is less than 18 inches. At the convent of Hanle in Tibet at 
the same elevation the pressure is 1 7 inches. 

Analogies are made with the experiences of men working under 
high pressure in caissons. The use of a lockout chamber or room, 
in which a man is allowed to become accustomed to increasing 
pressure, suggests that mountain climbers should become gradu- 
ally acclimated to greater altitudes. This was done in the case of 
the climbers who attacked Mount Everest. 

Balloonists have complained of symptoms the same as those 
experienced in mountain sickness. Since the preponderance of 
oxygen is in the lower atmosphere, this might indicate that sup- 
ply of oxygen is the most important factor to be considered. 



194 MOUNTAIN GEOGRAPHY 

However, in 1875, two French balloonists who reached 6939 
meters died with oxygen still in reserve, so that death could not 
have been due to lack of oxygen. The fact that the mouths of 
both men were clogged with blood caused many to believe that 
death was due to the effects of the sudden change in pressure. 
Others believe their deaths to have been due to suffocation by 
gas released in the rapid rise of the balloon. More recently Ger- 
man balloonists in an open basket have survived 10,980 meters. 

The difference in pressure, or the lack of oxygen, no doubt 
accounts for many of the symptoms of mountain sickness, but 
once again stress may also be laid on the similarity existing be- 
tween the symptoms of mountain sickness and those of ordinary 
fatigue. It is known that excessive fatigue produces a tired feeling 
in the limbs, rapid, feeble, and irregular pulse, rise of tempera- 
ture, and vomiting. These are all symptoms common to moun- 
tain sickness. Mountain ascents entail an amount of work the 
extent of which is not always recognized. Exertion is inseparably 
bound up with consumption of oxygen, and since there is less 
oxygen at the high levels fatigue will declare itself sooner there. 
However, early beliefs of mountaineers that bodily weakness in- 
creases more or less regularly with increasing altitude have been 
somewhat modified. All other things being equal, the higher one 
goes, the less should be the effect of any given rise, since atmos- 
pheric pressure diminishes less rapidly as one ascends. Conse- 
quently the difference in effort required between one stage and 
another should be less with each succeeding stage of altitude. 
Mountain climbing exhausts not only through physical effort but 
also through the mental anxiety due to the dangers of the ascent. 
The inability to carry any but concentrated foods, and the fact 
that cooking by boiling becomes increasingly difficult with alti- 
tude, result in an inadequate diet. A disordered stomach enor- 
mously decreases one's power of resistance. Hill men of Asia chew 
clove and ginger while crossing high elevations in order to avoid 
the danger of indigestion. 

Hepburn goes so far as to reach a conclusion that mountain 
sickness experienced below a certain height (probably 5000 me- 
ters) is due largely to fatigue alone. The fact that practised 
climbers do not complain so much about mountain sickness as 
the casual visitor is due, he believes, to their better preparation 
for the ascent. On Aconcagua in Chile (7010 meters) an alpinist 
reports that his party suffered much at various times at altitudes 




Fig. 56. Winter Village in the Valley of the Plessur, Grisons 














Fio. 57. Summer Village in the Valley of the Sapunerbach, Grisons 
This valley is a tributary of the valley of the Plessur pictured in Figure 56. 



POPULATIONS AND DISTRIBUTION 1 95 

over 4900 meters. In Kashmir the people who live at an eleva- 
tion of 1800 meters often suffer from mountain sickness after 
passing 3000 meters. Whymper and his companions on the 
slopes of Chimborazo were affected at 5100 meters, but Wolf and 
Whymper did not suffer at all on the summit of Cotopaxi at 
5960 meters, although they spent a night at this altitude. 

There is much disagreement as to the degree to which man can 
acclimatize himself to the higher altitudes. The conclusion that 
after a period of acclimatization man can live at the higher alti- 
tudes without a great deal of discomfort is no doubt true. Ex- 
perience tends to show that with a sufficiently slow change of 
altitude, allowing time for the body to adjust itself to changes of 
oxygen and pressure, mountain sickness can be escaped. 

Indians bringing sulphur from the summit of Popocatepetl, 
living at an altitude of 4000 to 5000 meters, seem strong and 
healthy after being engaged at the occupation for from twenty to 
thirty years. The experience of workmen engaged in the con- 
struction of a railroad tunnel in Peru at the height of 1 5,645 feet 
was that after several weeks, in extraordinary cases after several 
months, acclimatization was so complete that the men were ca- 
pable of doing full work. Many people, mostly railway men, living 
and working at altitudes of fourteen or fifteen thousand feet along 
the Railway of Southern Peru have never experienced mountain 
sickness. On the other hand, acclimatization to high altitudes 
over a period of generations no doubt takes the form of natural 
selection of the most fit. The Bolivian Indian with his deep chest 
and great endurance, living at altitudes of 8000 to 14,000 feet, is 
a good example. 

Since mountain sickness is influenced by a variety of things, 
such as the human constitution, habits, exertion, and the attend- 
ant natural circumstances of relief, pressure, wind, weather, and 
so forth, it is not surprising that the altitude at which the sickness 
appears varies greatly. 

Among the subjects which involve the relation of altitude and 
man is the therapeutic influence of insolation at high altitudes. 
So great is its importance that sanitariums, solariums, and hotels 
concentrated at favorable positions have created towns at high 
altitudes. There is, indeed, a considerable population above the 
winter haze. 

This haze over valleys is a feature worth mentioning. Peaks 
and the flanking alps have greater cloudiness in summer than do 



196 MOUNTAIN GEOGRAPHY 

valleys. But the moisture of valleys in winter rises only a short 
distance before it is condensed. Ordinarily by the time the alps 
are reached the haze has been passed. The writer recalls starting 
from Montreux on a December day. The air was cold but the 
skies were 'clear. 5 Upon arriving by funicular at the Chesieres 
Alp the light was dazzling, and the valley below was seen through 
a heavy haze. The ' clear' sky as seen from below was in reality a 
haze. Air temperatures on the alp were so high that a topcoat 
was a discomfort. Children from the solariums and health 
schools were skating or skiing with shoes, stockings, and abbrevi- 
ated trunks as their only clothing. 

The intensity of insolation has been treated in the foregoing 
pages. All that need be said of the intensity here is the contrast 
between winter intensities in valley and on alp. 

The significant aspect of the therapy of sunshine at high levels 
is the degree to which the short wave length ultra-violet rays 
come through unabsorbed. The importance of rays from the 
' finer' end of the spectrum is a matter for the medical derma- 
tologist rather than the geographer. The curative values are 
attested by the number of famous sanitariums and institutes for 
medical research at high altitudes. Davos, Aroso, Leysier, Mon- 
tana, and Andermatt are famous the world over as health resorts. 
There seems to be under the stimulating sunshine an increase of 
red-blood corpuscles, healing of respiratory organs, reduction of 
tumors, building up of nerves, invigoration of the heart, cure of 
rickets, and particularly the cure of tuberculosis. 

The health resorts are ordinarily situated in high valleys which 
are not too enclosed. As many hours of sunlight as possible is the 
desired thing. A notable character of the sun in these high alti- 
tudes is its intensity from the moment of sunrise. Here, amid 
dramatic splendor, the sick recover in the rays of a beneficent 
sunshine at an elevation greater than Snowdon, the highest peak 
of South Britain. This is at a time when invalids in London and 
Berlin lie in rooms darkened by fog. 

Not only the sick, but thousands of healthy people, seek the 
higher resorts in winter for sport and sunshine. The electric 
funiculars bear thousands each week-end to high resorts, the most 
famous of which is St. Moritz in the Engadine. The beauty of the 
Alpine scene is no less important than the sunshine. The spiritual 
values are not less than the physical renewal. 

For sick or for well, there is a large group of permanent in- 



POPULATIONS AND DISTRIBUTION 197 

habitants who minister to their wants. Cities exist above the 
winter clouds. Sunshine, particularly ultra-violet sunshine, is 
responsible for the greatest anomaly in the altitudinal distribu- 
tion of man. 

Some of the curative properties of insolation at high altitudes 
are: 

It adds vitamin A to certain foods. 

It increases the calcium and phosphorus content in the milk of alp 
pasture cows as against the high citric acid content of milk from cows 
of the lowlands or fed on dry fodder. 

It increases phosphorus in human blood. A few minutes of body ex- 
posure each day is said to double the phosphorus content in a baby's 
blood. 

The thyroid gland is richer in the secretion of iodine. 

Human mechanical energy is increased. 

It cures rickets and lung diseases. 

It is generally detrimental to germ life. 



BIBLIOGRAPHICAL NOTES 

Mountain Population 

All regional, economic, and historical references on mountains are 
likely to contain information on population. 

Allix, Andre. UOisans au Moyen-Age. Paris, 1929. An example of the 
many studies in historical regional economy which treat of popula- 
tion. 

Arbos, Philippe. "Evolution 6conomique et d£mographique des Alpes 
francaises du Sud," in Comit6 des travaux historiques et scienti- 
fiques, Bulletin de la section de geographies xxviii (191 3), pp. 296-306. 
Another typical historical-economic study. 

Bauer, Andra. Entvblkerung und Existenzverhaltnisse in Vorarlberger Ber- 
glagen. Bregenz, 1930. 

Bernhard, Hans, Roller, Albert, and Caflisch, C. Studien zur Gebirgsent- 
vblkerung. Bern, 1928. (Beitrage zur Agrar geographies 4.) Excellent. 

Bowman, Isaiah. "The Distribution of Population in Bolivia," in 
Bulletin of the Geographical Society of Philadelphia, vii (1909), pp. 28-47. 

Clerget, Pierre. " Le peuplement de la Suisse," in Bulletin de la Societe 
royale beige de geographies xxx (1906), pp. 73-97. 

Corcelle, J. "La d6population des Alpes," in La nature^ xxxi, 1 (1903), 
pp. 188-190. 

Dainelli, Giotto. "La Distribuzione della Popolazione in Toscana," in 



198 MOUNTAIN GEOGRAPHY 

Memorie Geogrqfiche, supplement alia Rivista Geografica Italiana, xi 

09i7)> PP- 3- 26 °- 
Edwards, K. G. "Valley Settlement in North Tirol," in Geography, xvi 

(i93 ! )> PP- 197-206. 
Fabre, L. A. V evasion contemporaine des montagnards francais. Nancy, 

191 !• 

Fliiekiger, Otto. "Die obere Grenze der menschlichen Siedelungen in 

der Sehweiz," in £eitschrift fur schweizerische Statistik, 1906, i. Band, 

pp. 145-162. 
Folliasson, Mme. J. " Mouvement de la population en Maurienne au 

XIX e siecle," in Recueil des travaux de Vlnstitut de geographie alpine, iv 

(1916), pp. 1-187. 
Gex, Francois. "La population de la Savoie de 1921 a 1926," in Revue 

de geographie alpine^ xvi (1928), pp. 221-250. 
Krebs, Norbert. "Die Verteilung der Kulturen und die Volksdichte in 

den osterreichischen Alpen," in Alitteilungen der Geographischen 

Gesellschajl in Wien, lv (191 2), pp. 243-303. 
Letonnelier, G. " L'emigration des Savoyards," in Revue de geographie 

alpine, viii (1920), pp. 541-584. 
Lugeon, Maurice. "Quelque mots sur le groupement de la population 

dans le Valais," Etrennes helvetiques, 1909. 
Meuriot, Paul. La population de la Savoie par altitude. Paris, 1907. 
Montbas, Hugues. " Le peuplement des Alpes suisses: Sa repartition et 

ses limites d'altitude," in Alemoires de la Socicte jribourgeoise des sciences 

naturelles, serie geologie et geographie, viii (191 9), pp. 155-240. 

Also printed separately, Fribourg, 191 9. 
Peattie, Roderick. "Notes sur les populations des montagnes curope- 

ennes," in Annales de geographie, xl (1931), pp. 386-395. 
Perriaux, L. "L'agonie et la mort des villages des Alpes nic,oises: Dif- 

ferents processus dans le passe et dans le present," in Bulletin de 

V Association de Geographes francais, no. 75 (1934), pp. 6-7. 
Privat-Deschanel, Paul. "The Influence of Geography on the Distribu- 
tion of the Population of Scotland," in Scottish Geographical Magazine, 

xviii (1902), pp. 577~5 8 7- 
Reynier, E. "La region Privadoise," in Recueil des travaux de Vlnstitut de 

geographie alpine, iii (1915), pp. 1-56. An example of a region with so 

meager an economy as to have a slight population. 
Robert, Jean. " La densite de population des Alpes francaises d'apres la 

denombrement de 191 1," in Revue de geographie alpine, viii (1920), 

pp. 5-124. 
Sauvan, E. "L'evolution 6conomique du Haut-Diois," in Revue de 

geographie alpine, ix (1921), pp. 521-624. A changing economy. 
Sion, Jules. " Le Tibet meridional et Pexpedition anglaise a Lhassa," in 

Annales de geographie, xvi (1907), pp. 31-45. 
Toniolo, A. R. "Per uno Studio sistematico sullo spopolamento della 



POPULATIONS AND DISTRIBUTION 199 

Vallate Alpine," in Atti delV XI Congresso Geografico Italiano, ii (Naples, 

1930). Statistical. 
Toniolo, A. R. "Ricerche di Antropogeografia nelPalta Valca- 

monica," in Memorie Geografiche, supplemento alia Rivista Geogrqfica 

Italiana, vii (191 3), pp. 245-363. Reviewed at length by Philippe 

Arbos in Recueil des travaux de VInstitut de geographie alpine, iv (1916), 

pp. 259-267. A complete study. 
Toniolo, A. R. "Lo spopolamento montano nella Venezia Triden- 

tina," in Bollettino della Reale Societa Geografica Italiana, serie vi, viii 

O93O, PP- 99-1 ii- 
Wopfner, Hermann. " Eine siedlungs- und volkskundliche Wanderung 
durch Villgraten," in J^eitschrift des Deutschen und Osterreichischen 
Alpenvereins, lxii (1931), pp. 246—276. 

Locations of Dwellings and Settlements 

Allix, Andre. "La limite superieure de Phabitat permanent dans les 

Alpes," in Revue de geographie alpine, xi (1923), pp. 293-297. 
Allix, Andre. "Note sur la valeur demographique du 'feu' dans 

Pancien DauphineY' in Revue de geographie alpine, xi (1923), pp. 635- 

640. 
Arbos, Philippe. "Etudes sur Phabitat de montagne en Italie," in Re- 
cueil des travaux de V Institut de geographie alpine, iv (191 6), pp. 259-274. 
Blanchard, Raoul. " L'habitation en Queyras," in La geographie, xix 

(1909), pp. 15-44, 97-110. An important article. 
Blanchard, Raoul. "Altitudes d'habitat," in Recueil des travaux de 

VInstitut de geographie alpine, vii (19 19), pp. 691-702. 
Chatelard, Maurice. "Les ph6nomenes d'habitat dans les Pyrenees 

Ariegeoises," in Revue geographique des Pyrenees et du Sud-Ouest, ii 

0930> PP- 44 8 -5!3- Analytical. 
Dellenbach, Marguerite E. "La conquete du Massif alpin et de ses 

abords par les populations prehistoriques," in Revue de geographie 

alpine, xxiii (i935)> PP- H?"^- 
Feuerstein, Arnold. "Damuls, die hochste standige Siedlung im Bre- 

genzerwald," in Geographischer Jahresbericht aus Osterreich, xiv-xv 

(1929), pp. 1-28. An excellent local study. 
Fliickiger, Otto. Die obere Grenze der menschlichen Siedelungen in der 

Schweiz- Bern, 1906. The great study on this subject. 
Fliickiger, Otto. " Passe und Grenzen," in Mitteilungen der Geographisch- 

Ethnographischen Gesellschaft in Zurich, xxvii-xxviii (1926-28), pp. 36- 

65- 

Gasperi, G. B. de. "Le casere de Friuli," in Memorie Geografiche, supple- 
mento alia Rivista Geografica Italiana, viii (19 14), pp. 295-461. Re- 
viewed at length by Arbos, in Recueil des travaux de VInstitut de geo- 
graphie alpine, iv (1916), pp. 267-274. 



200 MOUNTAIN GEOGRAPHY 

Knecht, Theodor. Siedlungsgeographie des Berchtesgadener Landes. Bad 

Reichenhall, 191 3. 
Lehmann, Otto. "Der Begriff der oberen Siedlungsgrenze, seine 

Herkunft, seine Bestimmungsmethoden und sein geographischer 

Wert," in Mitteilungen der Geographischen Gesellschqft in Wien^ lvi 

( I 9 I 3)> PP* 33 2 ~394* An academic discussion reviewing previous 

literature. 
Lehmann, Otto. "Aufruf zur Einsendung von kurzen Mitteilungen 

uber hochste standige Wohnsitze in den Ostalpen mit Einleitung 

und Begrundung," in Mitteilungen der Geographischen Gesellschqft in 

Wien, lxii (191 9), pp. 345-359. Factors in height limits and methods 

for collecting data. 
Lehmann, Otto. "Die Besiedlung und die Verkehrstrassen," in Die 

bsterreichischen Alpen, ed. by Hans Leitmeier (Leipzig and Vienna, 

1928), pp. 1256-287. 
Lehmann, Otto. "Fortbildung des Begriffes der oberen Grenze der 

Dauersiedlungen in den Alpen," in Mitteilungen der Geographischen 

Gesellschaft in Wien, lxiii (1920), pp. 153-162. 
Lowl, Ferdinand. " Siedlungsarten in den Hochalpen," in Forschungen 

Zur deutschen Landes- und Volkskunde, ii (1888), pp. 399-449. Detailed 

information. 
Lugeon, Maurice. " Quelques mots sur le groupement de la population 

du Valais," in Etrennes helv eti que s pour igo2 (Lausanne, 1902). 
Marinelli, Olinto. " I limite altimetrici in Comelico," in Memorie 

Geografiche, supplemento allaRivista Geografica Italiana, i (1907), pp. 9-99. 
Merlini, Giovanni. " I centri abitati della Garfagnana in rapporto al 

terreno," in Bollettino della Reale Societa Geografica Italiana, serie vi, 

vii (i93°)>PP- 29-49- 

Montbas, Hugues. " Le peuplement des Alpes suisses: Sa repartition et 
ses limites d'altitude," in Memoires de la Societe fribourgeoise des sciences 
naturelles, serie g£ologie et geographie, viii (19 19), pp. 155-240. Also 
printed separately, Fribourg, 191 9. In spite of the title this study 
is confined to areas above 1000 meters. This, with the study by 
Fliickiger, covers the subject. 

Osterreichische Ingenieur- und Architekten-Verein. Das Bauer nhaus in 
Osterreich-Ungarn und in seinen Grenzegebieten. Dresden, 1906. 

Ott, Adolf. "Die Siedelungs-Verhaltnisse beider Appenzell, ,, in 
Jahresbericht der Geographisch-Ethnographischen Gesellschaft in Zurich, 

xiv-xv (i9i3- x 5)> PP- 33" l6 3- 

Reishauer, Hermann. "Siedlungen der Deutschen und Italianer im 
Gebiete der Ostalpen," in Z u Friedrich Ratzels Geddchtnis (Leipzig, 
1904), pp. 289-302. 

Rinaldini, Bettina. "Die Obergrenze der Dauersiedlung und die rela- 
tive Hohe des Siedlungsraumes in Tirol," in Mitteilungen der Geo- 
graphischen Gesellschaft in Wien, lxxii (1929), pp. 23-47. 



POPULATIONS AND DISTRIBUTION 201 

Sieger, Robert. " Zur Geographic der zeitweise bewohnten Siedlungen 
in den Alpen," in Verhandlungen des Sechszehnten Deutschen Geographen- 
tages zu Number g (Berlin, 1907), also in Geographische JZ<ritschrift 9 xiii 

(i907) 5 PP- 361-369. 
Taylor, Griffith. "Settlement Zones of the Sierra Nevada de Santa 

Marta, Colombia," in Geographical Review, xxi (1931), pp. 539-558. 
Wopfner, Hermann. "Die Besiedlung unserer Hochgebirgstaler," in 

Zeitschrift des Deutschen und Osterreichischen Alpenvereins, li (1920), pp. 

25-86. 

Altitude and Health 

The many controversial aspects of this subject have brought out a 
multitude of writings. Many Alpinists have contributed to the subject. 
Hann, ed. by Knoch, i, pp. 210-227, has numerous references in the 
footnotes. The Verhandlungen der Klimatologischen Tagung in Davos, 1925, 
contains some forty articles upon the medical character of high altitude 
climates. Switzerland has led in mountain therapy and in the Davos 
volume there is perhaps the most important collection of opinions 
upon such matters that can be found in any single reference. 

Bosanquet, R. H. M. "Mountain-Sickness; and Power and Endur- 
ance," in Philosophical Magazine, 5th series, xxxv (1893), pp. 47-52. 

Conway, W. M. Climbing and Exploration in the Karakorom-Himalayas, 
London, 1894. 

Dent, Clinton. " Physiological Effects of High Altitudes," in Geographi- 
cal Journal, i (1893), pp. 46-48. 

Dorno, Carl. Studie iiber Licht und Luft des Hochgebirges. Braunschweig, 
191 1 . Dorno is here an authority. See his complete bibliography in 
the Bulletin of the American Meteorological Society, xiv (1933), pp. 282- 
286. See also his "Papers on the Relation of the Atmosphere to 
Human Comfort," in the Monthly Weather Review, liv (1926), pp. 39- 

43- 

Greim, G. "Der Mensch auf den Hochalpen: Nach Mosso," in Geogra- 
phische geitschrift, v (1899), pp. 94-104. 

Griffith, George. " Mountain Sickness," in Mature (London), Hi (1895), 
p. 414. 

Hepburn, M. L. "The Influence of High Altitudes in Mountaineer- 
ing," in Alpine Journal, xx (1901), pp. 368-393. A review which is 
most important. 

Jourdanet, Denis. Influence de la pression de Vair sur la vie de Vhomme: 
Climats d? altitude et climats de montagne. Paris, 1875. 2 vols. 

Kellas, A. M. "A Consideration of the Possibility of Ascending the 
Loftier Himalaya," in Geographical Journal, xlix (191 7), pp. 26-48. 
Physiological considerations. One should follow this with articles 
in the same journal on the attempts at Mount Everest. 



202 MOUNTAIN GEOGRAPHY 

Latham, D. V. "Kilimanjaro and some Observations on the Physiol- 
ogy of High Altitudes in the Tropics," in Geographical Journal, lxviii 
(1926), pp. 492-505. 

Longstaff, T. G. Mountain Sickness and its Probable Causes. London, 
1906. 

Mosso, Angelo. Fisiologia delV uomo sulle Alpi: Studiifatti sul Monte Rosa. 
Milan, 1897. 

Mosso, Angelo. Der Mensch auf den Hochalpen. Leipzig, 1899. An excel- 
lent scientific work. 

Whymper, Edward. Travels amongst the Great Andes of the Equator. New 
York, 1892. See pp. 366-384, and the appendix. Whymper was a 
great Alpinist. 

Workman, W. H. "Some Altitude Effects at Camps above Twenty 
Thousand Feet," in Appalachian xi (1905-08), pp. 350-359. 

Zuntz, Nathan; Muller, Franz; Gaspari, W. Hbhenklima und Bergwan- 
derungen in ihrer Wirkung auf den Menschen. Berlin, 1906. 
The Transactions of the American Climatological and Clinical Asso- 
ciation contain much material on this subject. 



CHAPTER IX 
MATTERS POLITICAL 

Mountain States 

MOUNTAIN regions and mountain valleys within these re- 
tions tend to establish, preserve, and increase political in- 
dependence. The mountain masses or the valleys, where they 
are sufficiently set off from the surrounding territories, become 
separate political entities, such as communes, cantons, or inde- 
pendent states. Where independence does not exist there are 
frequently movements toward separation. Where a degree of in- 
dependence has been won, the isolation and defensibility of the 
mountain fastness aid in resisting encroachment or maintaining 
autonomy. Thus the mountain mass which we know as Grisons 
was for a long time quite independent of the Swiss Confederation. 
Was it not the Asturias in the Cantabrian Mountains which alone 
of the Iberian provinces never bowed to the Moors? The moun- 
tain Kurds between Turkey and Persia owe little allegiance to 
either power, and in the isolated mountains of the Chinese plain 
there have always been semi-independent peoples. 

Tibet is the largest isolated political mountain unit of the 
world. It rests on the ' roof of the world ' at an elevation of some 
4600 meters. The great Himalayas lie to the south, and to the 
north are the formidable Kuen Luen Mountains, where passes 
leading to Turkestan are seldom less than 5000 meters in altitude. 
The access from the east has been easier than from the north or 
south, and hence the cultural and political allegiance has been 
Chinese. Today, remarkable as it may seem, Tibet is dominated 
politically by English influence, which has penetrated through 
the most forbidding range in the world. Actually the life of 
Tibet flows on with little regard for China or England, and the 
country will always remain politically aloof. 

The Himalayas are of so great a height, the valleys are so 
deeply inset, and the passes so difficult, that that great range 
holds other completely defined political entities. Nepal is a na- 
tive Indian state of which very little is known because of the ex- 
treme antipathy of its people toward the foreigner. Nepal is set 



204 MOUNTAIN GEOGRAPHY 

deep in the Himalayas between Tibet and India. Its area is 
50,000 square miles, and it has a population of some five millions. 
Physical unity has made Nepal a homogeneous state in spite of 
racial differences. The country has always been hostile to 
Chinese domination, and Great Britain duly recognizes Nepal's 
growing military strength, and draws thence many fine soldiers, 
the Ghurkas. Bhutan also lies between Tibet and British India. 
It has long persisted as an independent state despite the pressure 
of the great commercial empire to the south. Kashmir is the best 
defined of the Indian states. It is in one of the most impressive 
gorges in the world. No other large valley is flanked by more 
overpowering mountain walls. 

A still more isolated and mountain-guarded political entity, 
until its conquest by Kashmir in 1834-42, was Ladakh, now a 
district of Kashmir at the very headwaters of the Indus. It is the 
highest inhabited country in the world. The peaks that define 
its territory rise to 5800 meters. Leh, the capital, at 3355 me- 
ters, has the highest meteorological observatory in Asia. The 
failure of so isolated an area to maintain political independence 
came from the fact that its scant resources and population were 
insufficient to resist the overwhelming power of Kashmir. 

Europe likewise has mountain states. Montenegro in the Bal- 
kans, now part of Yugoslavia, proudly maintained its inde- 
pendence until the recent incorporation. Albania, a chaos of 
mountains and plains, has resisted not only effective political 
domination but modern culture as well, at least until the infiltra- 
tion of Italian commerce following the World War. The hill re- 
public of San Marino has been independent since the Middle 
Ages. More hidden among mountains is Andorra, where inde- 
pendence of her powerful neighbors is taught to the young at 
every hearthstone of the little country. 

The Swiss Confederation is born of the mountains. Even 
though today most of the Swiss live upon the plain, that plain is 
marked on the north by the Jura Mountains. But Swiss inde- 
pendence is an outgrowth of her isolated valleys rather than of 
the mountain mass as a whole. The original Swiss Confedera- 
tion was cradled in the valleys of four mountain forest cantons in 
the opposition to the tyranny of the Hapsburgs. The definite 
physical boundaries of the canton of Grisons nurtured the spirit 
of freedom which established its federal peasant-republic. 
Within this little republic there were three distinct leagues set 



MATTERS POLITICAL 205 

apart by natural barriers. For a time this republican federation 
of federations was in alliance with the Swiss Confederation; it did 
not become an integral part of Switzerland until the Treaty of 
Vienna in 181 5. The pastoral republic of Appenzell, independ- 
ent from about 1400, entered into alliance with the federated 
Swiss cantons in 1 452 and was admitted into the confederation in 

I513 ' 

The Austrian mountain provinces differ in culture and play 

distinct roles in the nation of today. The Hohe Tauern range 
separates the provinces of Salzburg and Carinthia. These two 
provinces, while they can communicate with Tirol, cannot easily 
communicate with each other on account of their mountain bar- 
riers. Since the defeat of Austria in the Great War, there has 
been a separatist movement in Vorarlberg. 

The autonomy of mountain districts is due to physical inac- 
cessibility, differences of climate, and the economic contrasts be- 
tween plain and mountain. It is, however, to be noted that the 
meager resources of the mountain as compared with those of the 
plain have often permitted plains people to rule mountains. 
Witness the Roman conquest of the Alps in the Augustan Age 
and the Japanese control of the mountain folk, to cite two ex- 
amples widely apart in time and space. Nevertheless it has rarely 
been worth while to subjugate mountains so thoroughly as to 
extirpate their peculiar spirit and culture. 

Mountains As Boundaries 

The influence of mountains on political history has varied at 
different times, but, on the whole, it is true to say that next to the 
sea itself there has been no greater element controlling the settle- 
ment and progress of races and the political institutions of man- 
kind. Mountains are a natural element, permanently fixed, and 
they are at the same time a wall of protection to one people and 
a barrier of defiance against another. We may read little of the 
Lazi, warders of the Caucasian Gates against the hordes of the 
steppes, yet few peoples in ancient history had a more important 
part to play than they. While, in a way, modern methods of 
communication and transportation, telephone and tunnel, have 
pierced the walls of the great mountain masses, we still have im- 
portant mountain boundaries. We cannot think of Italy without 
the Alps. 



206 



MOUNTAIN GEOGRAPHY 



Mountains have to a considerable extent determined the 
boundaries of European states. Notable among the great moun- 
tain boundaries of the world is the wall of the Pyrenees. No less 
than five states abut upon the Alps alone, while the southern 
highland frontiers of Germany are formed in turn by the Bava- 
rian Alps, the mountains of the Bohemian Forest, the Erzgebirge, 
and the Sudeten. The Carpathians bound Poland on the south 



A n dLor rcu 



n 



> 



Fig. 58. Boundaries of Andorra 

The stippled areas are portions of Andorra beyond the true drainage basin 
of the country. 



and Czechoslovakia on the northeast. More often than not moun- 
tains are effective and strategic frontiers. The effectiveness of 
mountain barriers depends upon their height, their lineal extent, 
their breadth, and particularly upon the accessibility of their 
passes. No less important has been the character of the relief. A 
single precipice of no great height may be as effective as the far 
greater height of the massif. Forested mountains are particularly 
effective as barriers; witness the low Bohemian Forest. 



MATTERS POLITICAL 207 

On the other hand, mountains as barriers to political expan- 
sion have in some cases been notably ineffective. An exception 
to the rule that mountain ranges form boundaries is seen in the 
Central Andes. It would be thought that so great and decisive a 
rock wall would everywhere be a political limitation, yet Ecua- 
dor, Bolivia, and Peru seem little concerned as to crest line or 
watershed. The Urals separated Russia from Siberia, yet they 
presented no effective barrier to Yermak and his Cossacks. Po- 
litical ambitions or desire for land incite nations to cross even 
great mountain ranges. 

Between Slovenia in Yugoslavia and Carinthia in Austria 
there is the Karawankcn range. This is a formidable Alpine 
chain with only a single difficult pass. One of the longer tunnels 
of the world today gives rail connection between the flanking 
plains. On the north side of the Karawankens is the basin of 
Klagenfurt. One would expect the basin of Klagcnfurt to be 
ethnically unlike the southern or Slovenian slope. As a matter of 
fact, the Slovenes occupy the southern part of the basin. Through 
a plebiscite it was made possible for the voters to set the political 
boundary either north of the line of the Slovene settlement, or 
north of the basin, or set it south of the basin. Fortunately the 
Slovenes of the basin voted in October, 1920, to join with Aus- 
tria. The expansion of the Slovenes had taken slight notice of the 
barrier effect of the Karawankens. On the other hand, the 
Slovenes north of the range have agreed to unite their economic 
future with that of Austria. Partly this was to avoid the domi- 
nation of the Serbs in Yugoslavia, but they also recognized the 
importance of the mountain barrier. 

We shall do well to look into the details of boundaries on some 
of the acknowledged barriers. Is the crest line of the mountain 
always a boundary? Is the watershed the boundary? 

The Pyrenees Mountains are frequently cited as an example of 
a mountain range whose watershed is a national boundary. In 
the Pyrenees, however, it is the chief mountain crest which is 
chosen, and watershed and chief mountain crest do not there 
always coincide. The exceptions to the coincidence of political 
boundary and watershed are many. We mention the two best 
known cases. The upper valley of the Garonne, the Val d'Aran, 
is Spanish. This valley, belonging to Spain, is reached from the 
south only by a mountain pass. Communication between the Val 
d'Aran and France is interrupted only by a senseless custom bar- 



qo8 MOUNTAIN GEOGRAPHY 

rier. The Segre River, a branch of the Spanish Ebro, rises on the 
plateau of Cerdagne well within France. 

Ordinarily one would think of a mountain valley as a single 
political area. Yet, because of morphological history, mountain 
rivers may flow through several mountain ranges. Thus the 
River Inn rises in Switzerland, where its valley is the well de- 
fined Engadine. It forms the valley of Central Western Austria. 
Yet lower in its course it drains part of the Bavarian plain. 

The question of the watershed as a political limit came to the 
front in the boundary dispute between the cantons of Valais and 
Bern in Switzerland. Bern demanded the fixing of the boundary 
at the watershed according to an established principle of inter- 
national law, while Valais took the stand that the limit should 
extend as far as the property rights of her citizens. Bern objected 
that such extension of property rights would prejudice her sover- 
eignty. Here there is a conflict of two geographical principles. 
The first is well substantiated. The water divide between two 
valley heads is the logical boundary for defense. The second is 
more economic than political. It involves the right of a people 
who lack pasture land to balance properly their holdings and so 
to extend their territory beyond a divide into an area of excess 
pasture. This type of situation is not uncommon in mountains. 
Figure 58 illustrates such a case in Andorra. 

There is yet another type of exception to the generalization 
that mountains make good boundaries. Mountains in certain 
economic matters are an area of transition rather than a sharp 
boundary. There is an economic interest in the zone along 
mountain heights which may override the strategic importance 
of the crest line. In peace economic interests tend to prevail, 
while in war the question of defense assumes the greater impor- 
tance. The balance of the two factors determines the location of 
the boundary. There is no question today of war between Valais 
and Bern. But in the case of the French-Spanish border or the 
Austrian- Italian border the case is different. Shall nations 
recognize the economic needs and claims of high mountain 
folk, or shall they insist upon the strategic boundary of the 
crest line? 

The case of the mountain folk is worth stating. Those people, 
whose way of life and economic interests are much the same on 
both flanks of the crest, have what we shall term 'straddle 
economies.' 



MATTERS POLITICAL 209 

On approaching an alpine, that is, glaciated range, one enters 
a valley flanked by mountain spurs. There has been no per- 
ceptible change in grade on leaving the plain and traversing the 
valley floor. The people in this valley, though living among the 
mountains, are hardly mountaineers. The chief agricultural in- 
terest is in the valley floor, and secondarily in the mountain 
spurs. Commercially the towns are marginal to the plains with 
the mountain back country as a tributary region. 

Farther up the valley are communities whose interests are 
first pastoral and secondarily agricultural. Agriculture is of the 
kitchen garden type and, in the past, a bread-stuff type. There 
is no raising of grains in excess of the local demand. Most fields 
are given over to the raising of hay for winter support of the herds 
and flocks. Upon the herds and flocks rests the economy of the 
community. The herds for a considerable part of the year are 
feeding upon alp meadows above the timber line. The ownership 
of the upland pasture is vital in the economic program. The men 
of these high mountain communities then face, economically 
speaking, up the mountain rather than down the valley. They 
are, in distinction to the man of the lower valley communities, 
mountaineers. Moreover, most great Alpine ranges have crystal- 
line cores. Their crests are not always well defined. The upland 
may be of the dome order. Such mountains have plateaulike 
surfaces, or, at least, broad, flattish saddles between the peaks. 

It is these upland areas, these skyline pastures, that are the 
physical basis of the political circumstance we are considering. 
Men of villages of opposing slopes have an equal concern for the 
pastures. In some cases, because of the asymmetry of the water- 
shed, a village may make use of pastures which arc on the far side 
of the crest line, as in the case of Valais. Villages not uncom- 
monly control pasture lands which are thus without the political 
domain. The sharing of these upland pastures by men of the 
opposing mountain slopes binds together the communities of the 
two slopes. There is an alpine culture, a high level economy, 
which straddles the mountain range and extends down until, in 
the valleys, the culture of the plains is met. Such high level cul- 
tures common to the opposing upper slopes of a range are c strad- 
dle economies. 5 Let us call peoples of such economic and political 
tendencies c straddle folk. 5 The recognition of a straddle economy 
implies a reduction of the significance of mountains as bound- 
aries, and is well illustrated by the Pyrenees. 



2IO MOUNTAIN GEOGRAPHY 

Straddle Economies of the High Pyrenees 

The Pyrenees are not so sharp a dividing line between France 
and Spain as they are generally supposed to be. At places along 
the summit of the range it is difficult to judge where the water 
divide may be. In many parts of the Pyrenees the French and 
Spanish summit pastures are more or less continuous. With pas- 
toral life as a prime element in the existence of both the Spanish 
and French mountaineers, these pastures are, in summer, the 
true focus of the local economies. The flocks and herds of the two 
people here mix. A bond, born of long contact with each other 
and a common set of economic problems, is established between 
the people of the two nations. Though more so in the past, even 
today the Pyrenees are a zone of transition rather than a sharp 
boundary between governments and cultures. Cultures straddle 
the range. The Catalans are found in almost equal strengths on 
the two slopes of the eastern end, as are the Basques at the west- 
ern. The French Bearnais of the Central Pyrenees speak a dia- 
lect intelligible to the Spanish mountaineers. 

So differentiated has this transitional zone been from the 
flanking plains, that the separatist tendency, common to moun- 
tain peoples, finds here excellent illustrations. We are reminded 
of Swiss history when we learn that there was in this range for 
three centuries a little known federation. The mountaineers, re- 
gardless of the national policies of Spain or France, maintained a 
'state' possessing frontiers, public law, and a political conscious- 
ness. This federation was bound together largely by common 
interests in the high pasture terrain. Had the range crest to a 
greater extent been serrated, the state might not have existed. 
Frequently communes on opposite slopes had more commercial 
and social relations with each other than with the plains to which 
they belonged nationally, though not always linguistically. The 
treaties between such communes, known as traites de lies et de 
passeries, were many, and involved rights to pasture, wood, water, 
and commerce. Rights to resources and privileges to trade were 
exchanged without reference to the national governments, and 
were retained even when the nations to which the communes be- 
longed were at war. Thus Bareges (French) and Bielsa (Spanish) 
agreed in 1 384 to continue friendly relations in case of war be- 
tween Aragon and England. Ossau, Aspe, and Baretous were 
neutral during the Hundred Years' War. The residents of the 




Fig. 59. Hay Transportation in the Pyrenees 
Many mountain fields are inaccessible to carts. 




. / ■'-"..•;.v. 



Fig. 60. Loggers in Andorra 
Breakfast at the edge of the forest zone preparatory to 'snaking' the logs to 

the? vallfv. 



MATTERS POLITICAL 211 

French valley of Baretous still pay on each July 1 2 a tribute of 
three heifers to their Spanish neighbors of the valley of Roncal. 
The presentation is followed by inspection of their common pas- 
tures and a feast. In the War of the Spanish Succession, the 
transmontane commercial relations were maintained regardless 
of the will of kings. Indeed, republican tendencies are common 
enough in these mountains, the natural results of agreements 
between communes. 

The government of Andorra is an example of the same histori- 
cal movement. Andorra is indeed merely a federation of villages. 
Its autonomy has arisen out of physical conditions and as an in- 
heritance of feudal rule. Its physical situation in the Pyrenees 
is such that it is difficult to decide whether it should be Spanish 
or French. Moreover in feudal days there were frequently in- 
terests which traversed the mountains. Foix, Catalonia, and 
Navarre all held fiefs on the far side of the range. It came about 
that Andorra was at the same time a fief of the Spanish bishop of 
Urgel and the French count of Foix. It was thus apanage. 1 That 
it should have attained this status was partly because its central 
position gave neither one overlord nor the other the advantage. 
Its independence today is a direct consequence of the parage, 
which is, in turn, an outgrowth of its midway position. 

An Altitudinal Provincialism: The Tirol 

Straddle economies are in a sense altitudinal provincialisms. 
It is not surprising that they should exist, but rather that they 
have had so little political recognition, especially in the light of 
their clearly defined position. Thus in Central Africa, about 
Kilimanjaro, there are many cattle-breeders on the plains; a little 
higher on the piedmont there are gardeners with small irrigated 
tracts; while in the open lands above 1800 meters there are 
nomads. There is a three-fold cultural stratification of folk in 
Central Asia. The steppe lands hold a scattered nomadic people. 
The piedmonts, using mountain streams for irrigation, are more 
densely populated by tillers of the soil, while occasional cities are 
to be found, towns of commercial and industrial nature. The 
high mountain slopes are occupied by the pastoral agriculturalist 
seeking a meager livelihood. 

A simple economic zoning is plainly marked in the case of the 

1 Pareage, a fief held jointly by two overlords. 



212 MOUNTAIN GEOGRAPHY 

Tibetans of the western borderlands of Kansu. Below the 3000- 
meter level most of the inhabitants are farmers, but above that 
level they are nomads. Because of the changes of natural prod- 
ucts with altitude Tibetans do not always take mountain ranges 
or rivers for their boundaries. Sir Charles Bell describes his ex- 
perience. When on a tour of exploration through Bhutan to 
Tibet in 1904, he found that the boundary between these two 
countries at the point where both meet Sikkim was what the 
Tibetans called an ' upland-tree lowland-tree' boundary. The 
pine forests belonged to Tibet and the bamboo forests to Bhutan, 
which means, in effect, a contour of about 1 1,500 feet above sea- 
level. He goes on to say that this was a good practical boundary, 
for the Tibetans need the higher lands for grazing their yaks and 
upland sheep, while the Bhutans make great use of the bamboo. 
This is a boundary not easily recognized by Western people, who 
look for frontiers along high mountain ranges, which are easily 
defended and can be delineated on maps. 

The Tirol as a political term includes the area between the 
crest of the Bavarian Alps and the crest of the Austrian-Italian 
Alps in the region of the Brenner. In reality there are parts of 
three geographic regions within the Austrian province of Tirol. 
On the north, in the Bavarian Alps, is a culture which extends to 
the crests of the precipitous slope which marks the northern wall 
of the Inn Valley. The Inn Valley is characterized by valley 
farms, industries, and commerce. The slope south of the Inn has 
in its deep-set valleys true Tirolean culture. The Zillertal exem- 
plifies this distinctive and true Tirolean manner of life. This cul- 
ture is marked on the north more or less by elevation contours. 
Tirolean culture is characteristic of the high tributary valleys. 
Moreover, this culture in the region of the Brenner (Otztaler and 
Stubaitaler Alps) extends over the mountains into such deep 
gorges as Valle Passiria (Passeiertal), where life is as truly Tiro- 
lean as the highly characteristic life of the Zillertal. The high 
pastures of the massif were property of the people on the two 
flanks, and the relatively low Brenner offers the easiest transmon- 
tane communication in the length of the Alps. The Tiroleans on 
the south flank, that is, of the South Tirol, are pastoral people 
living distinctly as mountaineers. They are distinct in economy, 
social habits, and language from the grape-raising Italians. 
There is, then, on the south as on the north, an altitudinal 
differentiation. 



MATTERS POLITICAL 213 

Failure to recognize the South Tirol as an integral part of 
Tirol was one of the injustices of the Treaty of St.-Germain. 
Italy, remembering previous conflicts with Austria, hoped to ob- 
tain from the World War a strategic boundary which would ex- 
clude Austria from the southern slope of the Alps. President 
Wilson, seeing that the crest of the Alps was physically the logical 
boundary for Italy, lent his support to the Italian claim, but later 
bitterly regretted his action. At the end of the war the South 
Tirol became the Italian Trentino. This was an injustice in the 
light of race and culture which, to the Tiroleans, far outweighed 
the justice of granting Italy the new strategic boundary. The 
validity of the Italian claims can easily be seen if the point of view 
of military necessity alone be held. But Italy does not hold the 
crest line as a boundary against Switzerland. True, the moun- 
tain Swiss form a neutral state. Italy does, indeed, hold the crest 
line along the French border, and thanks to the folly of a French 
king, extends her boundary across the crest line so as to control 
strategically a large Provengal valley. The Italians are therefore 
not without precedent in their claims to the crest line. 

The fact remains that the Austrian South Tirol is today the 
Italian Trentino. German-speaking peoples are compelled by 
force to adjust themselves to the Italian regime. Tiroleans are 
serving in an army against which their elder brothers recently 
fought. It is forbidden in the taverns to sing German songs over 
the convivial glass. 

Importance of the Dalmatian Alps 

An excellent example of geographic principles in conflict with 
each other is exemplified by the problems associated with the 
Dalmatian Alps. These are a more or less continuous barrier 
extending from the Istrian Peninsula to the plateau of Mon- 
tenegro. The mountains are seldom truly alpine in character. 
They are a limestone upland, presenting an abrupt front to the 
littoral of the Adriatic and a more gradual slope with extended 
spurs towards the Yugoslavian interior. Were the plains of the 
interior more uniform the problem would be simpler. But the 
approach from the rich valley of the Save towards the alpine 
barrier is confused. 

The Dalmatian range so effectively cuts off the littoral from 
the interior as to form a coastal province whose people are dis- 



214 MOUNTAIN GEOGRAPHY 

tinct from the various interior peoples. There is a coastal Dal- 
matian people in distinction to the Carnilans, Slovenes, and 
Bosnians of the interior. The Dalmatians have a Venetian ele- 
ment in their historical culture and commonly are able to use 
Italian as a commercial language. 

This littoral is politically and commercially tied to the interior 
by the formation of Yugoslavia. The people have an intense 
national spirit which has stamped them as Yugoslavians rather 
than Italians. A geographic principle has been once again 
exemplified. A rich interior tends to expand so as to include a 
sea coast. So strongly has this principle asserted itself that the 
expansion has surmounted the mountains. Before the war the 
Save Valley had one communication by rail with the Adriatic. 
This was at Fiume. Fiume is now Italian and the Slavs are 
forced to use the suburb of Susak, unfavorable at best. Now, 
however, a railroad reaches the sea at Split and another ap- 
proaches Dubrovnik. 

There is a geographic principle of lesser force which asserts 
itself in this region. An enclosed sea is not unlike a peninsula. 
It tends to submit politically, commercially, or culturally to a 
dominant power. Rome early held both shores of the Adriatic 
under this principle, as did Venice at a later date. Italy has 
attempted to do likewise and today has several footholds 
upon the Dalmatian littoral. It holds Fiume, except for in- 
consequential Susak, the city of Zara, and certain islands. 
Farther south it plays the part of protector in Albanian politics 
and commerce. 

Italy holds Fiume and Zara on the basis of the extent of the 
Italian population there. If there were no Dalmatian Alps, if the 
plain of the Save merged with the littoral of the Adriatic, this 
aggression into Balkan territory would not have taken place. 
The principle of the expansion of an interior to include a littoral 
is ordinarily stronger than the principle of the unity of an en- 
closed sea. What has shifted the balance here is the barrier effect 
of the Dalmatian Alps. 

The barrier which mountains offer to political, cultural, or 
commercial aggression not only depends upon the physical char- 
acter of the range but upon the moment in history. Much de- 
pends upon the political, cultural, and commercial strengths of 
the forces on the two slopes at the periods in which the mountains 
and their human histories are studied. 



MATTERS POLITICAL 215 

Political Philosophies of Mountaineers 

Perhaps no side of the political geography of mountains has 
been more discussed than the political attitude of mountain folk. 
Many of the statements are broad generalities which cannot be 
substantiated. Yet there are political attributes of mountaineers, 
which, if imponderable, nevertheless deserve discussion. Ratzel, 
Miss Semple, Vallaux, and others have insisted upon a relation- 
ship between the state and the soil. In mountains, the size of the 
terrain and the character of the topography, as well as the degree 
of relief, have been little short of deterministic in matters eco- 
nomic and political. But have the inherent characteristics of the 
mountains affected the political theories of mountaineers? We 
appreciate the fact that soil, area, and climate have stamped 
upon Denmark's government the character of a cooperative 
dairying association. In the Middle Ages the same factors, plus 
a few others, made Denmark a headquarters of pirates. The 
strength of mercantilism in England is a function of insularity, 
over-population, and lack of agricultural back country. Do 
mountains suggest distinctive elements of political philosophy? 

The very difficulties and hardships incident to mountain life 
are a challenge to man's energies. Therefore, we find in the 
population of mountain regions a large measure of self-reliance 
and resourcefulness, together with a rugged hardihood. Moun- 
tains impose their character, as it were, upon the inhabitants 
whom they train. There is the fatigue of ascents and descents, 
the simple food, the constant struggle for existence: these have 
given the mountaineer a character all his own. He has usually 
been rewarded with liberty and other blessings that go with non- 
interference. In Switzerland we find not only political and re- 
ligious liberty, but the inevitable traits of character that develop 
from these privileges. Every man, peasant or workman, stands 
on his own feet. He is trained to respect leadership, but to de- 
spise and reject coercion. He is taught to believe that rights also 
involve duties in the exercise of citizenship. But if mountain 
peoples possess a commendable individualism and courage, as 
well as frugality, they must at the same time pay the penalties for 
their isolation. For the isolation of mountain regions not only 
prevents expansion and invasion, but also makes progress in ideas 
and inventions more difficult. 

That the mountaineer by the nature of his homeland is an in- 



2l6 MOUNTAIN GEOGRAPHY 

dividualist has long been a popular theme. It is not without a 
certain justification. The unbending individualism of the feud- 
waging mountaineer of the Southern Appalachians would sup- 
port the contention. Clans are confined to valleys. The valleys 
force upon their people an inbreeding, and thus it is the sepa- 
rated valley spaces rather than blood relationships which in the 
last analysis are the basis of the feuds. No organization yet de- 
veloped has broken down the social isolation of the mountain 
folk of the Balkans. The Balkans, the Scottish Highlands, and 
the Appalachian mountain areas are uplands maturely dissected. 
The social response to the character of dissection is in each case 
the same. Maturely dissected uplands are notorious for the feuds 
they engender. Their topography prevents progressive culture or 
social and political unification. 

On the other hand, the freedom of the mountaineer in alpine 
mountains is characteristic of the group rather than the in- 
dividual. In reality, the corridor character of great glaciated 
valleys, the communal character of the exploitation of alp pas- 
tures, the common danger from catastrophe, and the altitudinal 
distribution of property rights force upon the people of a single 
valley a cooperation which creates a syndicate with republican 
and socialistic aspects. People form a closely knit social group. 
Alp lands are owned in common because of their indivisible na- 
ture. Forested slopes are of public interest. Avalanches are to be 
prevented by community effort, by planting, or the construction 
of walls. Public routes of communication are rendered useless by 
floods, deposits, and avalanches, and must be restored by com- 
mon effort. Mountain terracing means detailed regulations con- 
trolling rights of access, erosion, and soil replacement. Irrigation 
on these terraces means the construction of a trunk canal and 
public regulation of the distribution of water. The sending of the 
cattle, sheep, and goats up to common pastures under a herds- 
man or group of herdsmen increases the group cooperation. The 
summer cheese industry means a proportional sharing of profits. 

In short, there is here a group organization, political and 
social, based upon a common interest in the combat against ca- 
tastrophes and in the use of resources. Nature forces upon the 
commune a group interest in resources and in the exploitation of 
those resources. The economic 'folk unity' is the basis of the 
political division. It has been asserted that from a sociological 
point of view there is no political boundary, but rather there is a 



4SKH 




Fig. 6i. Mounting to Potato Fields, Spanish Sierra Nevada 
The up journey takes four hours. Note the fuel — the men will remain over night 
in their fields, which are above the tree line. 




Fig. 62. Bringing Rye to the Village, French Pyrenees 



MATTERS POLITICAL 21 7 

shifting frontier. Such a statement has an element of truth for 
areas of the plains, but it has much less application to the moun- 
tain-rimmed area. 

Each area which has physical definition is found to support, to 
a greater or less degree, a group consciousness, a cultural pro- 
vincialism. 

There is, however, in mountains, an individualism of one val- 
ley as against another. How then is a confederation, such as 
Switzerland, formed from individualistic communes and can- 
tons? This is the more difficult to comprehend when one remem- 
bers that the valleys are radial and open upon the plains of several 
countries. The element of unity arises not so much from mutual 
commerce and interchange of resources and productions as from 
the common economic problems. A government of mountain 
valleys having the same economic and political problems is more 
rational than the government of mountain valleys by unsympa- 
thetic plains people. In their external relationships mountain 
groups have all the independence that writers have ascribed to 
the individual mountaineer. The confederation of valleys into a 
state is a matter of singleness of economic purpose and the need 
for protection. It is worthy of note that in Switzerland the com- 
munes are older than the cantons and the cantons existed before 
the confederation. The confederation was created, indeed, as a 
defense of communal and cantonal autonomy. 

The lack of continuity of level inhabitable land has played a 
role in political theory. This breaking up of men into small 
groups accounts for the fact that strong centralization of au- 
thority is not easily understood by the mountaineer. Mountains 
have, therefore, an essential suggestion of republicanism. The 
separate units in such a republic have their cultural and political 
views colored by the plains area in the direction in which their 
valley tends, and a mountain mass with radial valleys has a 
strong tendency towards decentralization of authority. Moun- 
tain areas may teach plains areas much in theories of liberty, 
but they will never be dynamic forces among nations because of 
their centrifugal interests. 



2l8 MOUNTAIN GEOGRAPHY 

BIBLIOGRAPHICAL NOTES 

The Political Geography of Mountains 

This list does not include treatises from the point of view of the politi- 
cal scientist, but rather works by geographers. 

Allen, W. E. D. "New Political Boundaries in the Caucasus," in Geo- 
graphical Journal, lxix (1927), pp. 430-441. 

Barnes, J. S. "The Future of the Albanian State," in Geographical Jour- 
nal, lii (1918), pp. 12-30. 

Bell, Sir Charles. Tibet, Past and Present. Oxford, 1924. 

Bishop, C. W. "The Geographical Factor in the Development of 
Chinese Civilization," in Geographical Review, xii (1922), pp. 19-41. 

Bowman, Isaiah. The Andes of Southern Peru. New York, 1916. 

Bowman, Isaiah. The New World. Yonkers, 1 92 1 . The great book in 
English on after-war political geography. 

Brigham, A. P. Geographical Influences in American History. Boston, 1903. 

Brigham, A. P. " Principles in the Determination of Boundaries," in 
Geographical Review, vii ( 1 9 1 9) , pp . 201-219. 

Brooks, R. C. Government and Politics of Switzerland. Yonkers, 191 8. 

Brunhes, Jean. Human Geography, tr. by I. C. Le Compte. Chicago, 
1920. 

Bryce, James. Modern Democracies. London, 1921. 2 vols. 

Cole, D. H. Elementary Imperial Military Geography. London, 1924. 

Coolidge, W. A. B. The Alps in Nature and History. New York, 1908. 

Corey, Herbert. "A Unique Republic, where Smuggling is an Indus- 
try," in National Geographic Magazine, xxxiii (19 18), pp. 279-299. 

Cowan, A. R. Master-Clues in World-History. London, 191 4. Chap. viii. 

Ehringhaus, Friedrich. Kleine Staatsbiirgerkunde, 4. Aufl. Gottingen, 

/925- 
Fairgrieve, James. Geography and World Power. London, 191 5. 
Fawcett, C. B. Frontiers. Oxford, 1921. 
Fliickiger, Otto. "Passe und Grenzen," in Mitteilungen der Geogra- 

phisch-Ethnographischen Gesellschaft in ^urich, xxvii-xxviii (1926-28), 

pp. 36-65. 
Freeman, E. A. The Historical Geography of Europe. London, 1881. 

2 vols. 
George, H. B. The Relations of Geography and History, 4th ed. Oxford, 

1910. 
Hauser, Henri. "La position geographique de la Suisse: Etude de 

g£ographie politique," in Annales de geographie, xxv (1916), pp. 413- 

428. 
Haushofer, Albrecht. Pass-Staaten in den Alpen. Berlin, 1928. 
Hettner, Alfred. Grundzuge der Landerkunde. Leipzig, 1907-25. 2 vols. 
Hogarth, D. G. The Nearer East. London, 1902. 



MATTERS POLITICAL 219 

Holdich, Sir Thomas. "Political Boundaries," in Scottish Geographical 
Magazine , xxxii (19 16), pp. 497-507. 

Huntington, Ellsworth. "The Vale of Kashmir," in Bulletin of the 
American Geographical Society, xxxviii (1906), pp. 657-682. 

Keller, A. G. Colonization, Boston, 1908. 

Little, A.J. The Far East. Oxford, 1905. 

Lyde, L. W. The Continent of Europe, 2d ed. London, 1924. 

Machatschek, Fritz. "Zur politischen Geographic der Schweiz," in 
Geographischer Jahresbericht aus Osterreich, xiv-xv (1929), pp. 1 15-135. 

Maull, Otto. Die bayerische Alpengrenze. Marburg, 1910. This, like the 
article by Penck, is detailed physical description of the character of 
the boundary. MaulPs article is of little value to the theorist. 

Newbigin, Marion Isabel. Geographical Aspects of Balkan Problems. Lon- 
don, 1915. 

Newbigin, Marion Isabel. The Mediterranean Lands. London, 1924. 

Newbigin, Marion Isabel. Southern Europe. London, 1932. Excellent. 

Ogilvie, A. G. Geography of the Central Andes. New York, 1922. 

Par tsch, Josef. Central Europe. London, 1903. Important. 

Peattie, Roderick. New College Geography. Boston, 1932. 

Penck, Albrecht. "Die osterreichische Alpengrenze," in ^eitschrift der 
Gesellschaft fiir Erdkunde zu Berlin, 191 5, pp. 329-368. 

"The Political Significance of Abyssinia," in Geographical Review, xiv 
(1924), pp. 147-148. An editorial note. 

Ratzel, Friedrich. Politische Geographic, 3. Aufl. Munich, 1923. 

Reclus, Elisee. The History of a Mountain, tr. from the French by Bertha 
Ness and John Lillie. London, 188 1. 

Ripley, W. Z. The Races of Europe. New York, 1899. 

Rohe, Alice. "Our Littlest Ally," in National Geographic Magazine, 
xxxiv (1918), PP. I39-I63- 

Semple, Ellen Churchill. The Geography of the Mediterranean Region. New 
York, 1 93 1. 

Semple, Ellen Churchill. Influences of Geographical Environment. New 
York, 191 1. Much information. 

Solch, Johann. "Die Ostalpen als geographischer Nachbar," in 
Zeitschrift fur Geopolitik, viii (1931)3 pp. 287-295. 

Tamaro, Attilio. La Venetie Julienne et la Dalmatie: Histoire de la nation 
Italienne sur ses frontier es orientates. Rome, 191 8-1 9. 3 vols. 

Tower, W. S. " The Andes as a Factor in South American Geography," 
in Journal of Geography, xv (191 6), pp. 1-8. 

Visher, S. S. "What Sort of International Boundary Is Best?" in Jour- 
nal of Geography, xxxi (1932), pp. 288-296. 

White, J. C. " Castles in the Air: Experiences and Journeys in Unknown 
Bhutan," in National Geographic Magazine, xxv (191 4), pp. 365-455. 

White, J. C. "Nepal: A Little-Known Kingdom," in National Geo- 
graphic Magazine, xxxviii (1920), pp. 245-283. 



220 MOUNTAIN GEOGRAPHY 

Political Geography of the Pyrenees 

Baring-Gould, Sabine. A Book of the Pyrenees. London, 1907. 

Brutails, J. A. La Coutume d*Andorre. Paris, 1904. The monumental 
book on Andorra. 

Carrier, Els6 Haydon. Water and Grass, London, 1932. Chap, xviii. 

Cavailles, Henri. "L' association pastorale dans les Pyr6n6es," in Le 
musee social , 1910, Memoires et documents, pp. 45-80. 

Cavailles, Henri. " Une f6d6ration pyr6neene sous Pancien regime: Les 
trails de lies et de passeries," in Revue historique 9 cv (19 10), pp. 1-34, 
241-276. 

Cavailles, Henri. " Notes sur les syndicats de communes dans les val- 
ines pyr6neennes," in Bulletin du Comite des travaux historiques et sci- 
entifiques, section des sciences economiques et sociales, Congres des 
societ6s savantes de 1908, pp. 193-201. 

Evans, E. E. "The Pyrenees: A Geographical Interpretation of their 
R61e in Human Times," in Studies in Regional Consciousness and En- 
vironment presented to H. J. Fleure, ed. by I. C. Peate (Oxford, 1930), 
pp. 45-68. 

Rios Urruti, Fernando de los. Vida e Instituciones del Pueblo de Andorra: 
Una Supervivencia Sehorial. Madrid, 1920. An excellent book, 

Whittlesey, Derwent. " Trans-Pyrenean Spain: The Vail d'Aran," in 
Scottish Geographical Magazine^ xlix (1933), pp. 217-228. 

South Tirol 

The question of South Tirol, or Trentino, is one of such importance 
that this brief bibliography is here given. It was prepared by Dr. Benno 
Graf of Munich from the Volksbund fur das Deutschtum im Ausland in 
a somewhat more complete form. 

Dorrenhaus, Fritz. Das deutsche Land an der Etsch. 1933. 

Fink, Hans. Die Kirchenpatrozinien Tirols. Passau, 1928. 

Haspinger. Wie Deutsch-Sudtirol von den Italienern behandelt wird. 1924. 

Hennersdorf, F. K. Siidtirol unter italienischer Herrschaft. Charlottenburg, 

1928. 
Mannhardt, J. W. Siidtirol: Ein Kampj urn deutsche Volkheit. Jena, 1928. 

Political. 
Reut-Nicolussi, Eduard. Tirol unterm BeiL Munich, 1928. Political. 
Rohmeder, Wilhelm. Das Deutschtum in Siidtirol. Berlin, 191 9 and 1932. 
Stolz, Otto. Die Ausbreitung des Deutschtums in Siidtirol im Lichte der 

Urkunden. Munich, 1927-34. 4 vols. 



CHAPTER X 
THE CHARACTER OF MOUNTAIN LIFE 

MOUNTAINS are found in all latitudes and all climates, 
set down amidst civilizations ranging from the most primi- 
tive to the most advanced. It is therefore impossible to write in 
general terms upon the social qualities of mountaineers. And 
yet in order to complete this volume it would seem necessary to 
describe some types of social problems peculiar to the men of the 
mountains. Briefly, the social traits which may be ascribed to 
mountaineers the world over are perhaps three: conservatism, 
arising from provincialism, frugality accompanied by low stand- 
ards of living, and an industry undaunted by an almost over- 
whelming load of toil. 

Mountain Conservatism: Appalachia 

There is a biological law which points out that any life form, 
isolated and not fed by new environmental influences and not 
stimulated by breeding from life forms outside the province, will 
not alter to the same degree as similar life forms subjected to 
changing environments and breeding with new varieties. Like- 
wise, customs and social attitudes in isolated mountain provinces 
with meager resources and few outside stimuli tend to crystallize. 
Mountain life is conservative. The Pyrenees, the Alps, and the 
Massif Central of France have contributed little directly to 
French civilization. The Appalachians, the Ozarks, and the 
Western Cordillera in the United States have made no progres- 
sive cultural contributions to American life. We find in the 
mountains of India the conservative temper in the most extreme 
form dominating the whole social structure. In Kashmir, in 
spite of its incorporation in the British empire, old ideas prevail 
to an extent which excludes new forms to a remarkable degree. 
Stein, an authority on the history of Kashmir, says the character 
of the masses of the people has changed but little for thirteen 
centuries. If this is true of the people of a great region like Kash- 
mir, we may expect to find smaller and more isolated communi- 
ties bound by tradition to a yet further extent. 



222 MOUNTAIN GEOGRAPHY 

Until the opening of the present century the mountain folk of 
the secluded valleys of the Pyrenees were traditionalists and hos- 
tile to innovations. This is less true of the valleys of the Alps, 
because they have been traversed by so many pilgrims and trav- 
elers passing between the northern countries and the Mediter- 
ranean world. But even within the Alps, out of the main lines of 
travel, certain separate valleys have maintained their conserva- 
tive ways of life. The Val d'Anniviers, high in the Valais, and 
the Engadine, remote and mountain-rimmed, are examples to 
prove the point. Physical isolation and meager resources have 
kept certain valleys almost medieval in character. It is difficult 
to conceive of greater mountain isolation than is found in valleys 
tributary to the River Var in Provence. Three branches of this 
river are from basins isolated from the outside world by almost 
impassable gorges. Hence each valley has maintained a distinct 
social life. The Vesubie tributary valley is well enclosed, and is 
made terrible by overhanging cliffs, earthquakes, and landslides. 
So forbidding is the valley that it is much shunned. There are 
reports of wandering evil spirits who are reincarnated Moors. 
At intervals the valley has been purified by a sprinkling of holy 
water. 

The difficulty of transportation in mountains isolates valleys 
from the stream of progress. This is true, likewise, of whole 
mountain masses in distinction to surrounding plains. Mountain 
masses are conservative to a degree of intensity directly propor- 
tional to the difficulties of inner transportation. The Aures 
Massif of Algeria is physically set off from the surrounding coun- 
try and has primitive facilities for transportation. The native 
customs are peculiarly local. This is shown by the construction 
of houses, the methods of irrigation, and the games, these last 
being a relic of the worship of Athena. The Jurdes are a people 
living in a group of low mountains in Central Spain. Sur- 
rounded by progressive folk, the Jurdes are literally savages. 
This small enclave of barbarism amidst civilization is a startling 
example of unprogressiveness. 

An American example of a mountain folk who suffer from 
isolation has been analyzed by John Wesley Coulter. The Santa 
Lucia Mountains lie along the coast of California south of Mon- 
terey. The mountaineers there are retarded economically by the 
ruggedness of the topography, the steep slopes, and the thin soils. 
Communication with the outside world has been limited. To 



THE CHARACTER OF MOUNTAIN LIFE 223 

reach markets with produce was almost prohibitively costly, and 
even where there was money the procuring of supplies was 
troublesome. 

This mountain range is in line with the north-south communi- 
cation through California, yet it has had no improved transpor- 
tation. One town, Lucia, was 25 miles (40 kilometers) by 
difficult mountain trails from a road on the north and an equal 
distance to a roadway on the south. The railroad station for the 
district is 40 miles (64 kilometers) away. This mountain group 
has many peaks more than 900 meters high. One peak attains an 
elevation of 1 782 meters. The difficulty of transportation lies in 
the narrow and steep-sided valleys and the poverty of passes. 
Lucia is indeed on the sea, but without adequate harbor. 

The result is that stock raising is prevalent because the beasts 
can walk to market. The journey from the Lucia region to rails 
ordinarily takes four days. The meager forage and the rough 
terrain produce an inferior cattle. A recent development has 
been the raising of swine. The hogs are transported by boat. 
The loading is done by means of an aerial cable. So difficult are 
the trails leading to the sea that a drove of hogs will make no 
more than a mile an hour. 

Supplies are brought to the region by launches that land on the 
beach. A trip to 'town' once a year is an event. An important 
occasion of a girl's life is that when she marries she is taken out of 
the mountains and given a ride in a wheeled vehicle. The com- 
ing of the mail carrier has more than usual interest. Education, 
religious development, and social intercourse correspond to the 
difficulty of transportation. A state highway is now being con- 
structed along the mountain-sides bordering the coast. The cost 
of such a road in ten-mile sections varies from $21,400 to $86,200 
a mile, but single miles in the worst locations may cost double the 
higher figure. 

Conservatism and the crystallization of culture are nowhere 
more easily observed than in the little world of the Appalachians. 
The Appalachian mountains and plateaus were barriers to the ex- 
pansion of the English colonists of the thirteen colonies. Gradu- 
ally the tidewater zone and then the piedmont became popu- 
lated. People began pushing through the mountain ranges. 
Some sought the mountain valleys by choice, while others, fa- 
tigued in their journey towards the Kentucky meadows, pushed 
no farther than the confused valleys of the plateau. Those who 



224 MOUNTAIN GEOGRAPHY 

went on and settled in the Blue Grass were distinguished in no 
way from the men who remained in the valleys except in per- 
severance and courage. Of those groups that cleared mountain 
farms by choice some were Scotch- Irish from Ulster and some 
Germans from the Rhenish Palatinate. Generally the settlers 
were of Anglo-Saxon stock. The ways of living introduced at the 
period of settlement have been altered by the local environment, 
but they have not been affected by stimuli from the progressive 
plains that flank the mountains. Nowhere in North America 
have communities been left so to themselves. The farmers served 
as blacksmith, gunsmith, carpenter, furniture-maker, and miller. 
Doctoring was accomplished with local herbs, and dentistry with 
a tool from the barn. 

At the beginning of the twentieth century the farm had a de- 
gree of economic independence which was remarkable. The 
cabin was of square hewn logs. The fireplace and chimney were 
of stone and clay. Corn meal and home grown vegetables, pork 
from the household pig, or squirrels shot in the woods constituted 
the food. The furniture was homemade and woolen and cotton 
cloth was homespun dyed with hickory. The household and im- 
pedimenta were colonial in character. The inhabitants have 
been called our * contemporary ancestors.' 

It is in the characteristics of their mental culture that pro- 
vincialism showed itself most strongly. In matters political the 
point of view is always local. A national election arouses them 
less than the choice of a local judge or sheriff. They have had no 
political alliance with the plains people, and, indeed, during the 
Civil War, they were opposed to slavery and secession. They 
have been individualists with tendencies towards separation. 
This is evidenced by the ' lost 5 State of Franklin which existed for 
a brief interval in Eastern Tennessee. 

Speech is the truest measure of the crystallization of culture. 
The Appalachian mountaineers speak an Anglo-Saxon which is 
as pure as exists today. Words obsolete in American diction are 
plentiful. 'Holp' and 'holpen 5 for 'help 5 and 'helped 5 are an- 
cient forms with an ancient ending. ' Poke 5 for ' bag' and ' buss' 
for 'kiss' are seldom found elsewhere in America. Their lan- 
guage is at times almost Shakespearean. In their community 
singings they still use old books with a form of musical notation 
elsewhere passed into museums. The ballads are those of seven- 
teenth-century England. Not so long ago a man was found 



THE CHARACTER OF MOUNTAIN LIFE 225 

hunting rabbits with an old English crossbow. This ingrowing of 
culture is paralleled by the physical inbreeding. In a stretch of 
forty miles along the Kentucky River every family has the same 
surname. 

Underlying the crystallization of culture, the individualism, 
and the almost complete isolation of these people is the mountain- 
ous terrain, the confusion of mature dissection of the plateau 
area, the lack of level land, the corridor valleys and passes, and 
the meager resources. 

Physical Factors in Standards of Living 

Since we have made so many generalizations involving altitude, 
we might add yet one more. In this case the object is to demon- 
strate the second social trait of mountain life, frugality accom- 
panied by low standards of living. Generally speaking, it is true 
that, in mountains of the temperate zones, the greater the altitude 
the lower the standard of living. This, of course, would not hold 
for elevations in the tropics. There has, indeed, been a serious 
suggestion that in Switzerland, in order that every one might 
have a proper minimum standard of living, those who live above 
a certain contour line have a subsidy, and that the subsidy be in- 
creased with altitude. This suggestion certainly supports the 
rather broad generalization which we have made. 

Altitude decreases the materials of life. High villages, unless 
mining camps or tourist centers, are universally poor and primi- 
tive. The rude climate precludes gainful agriculture. The alti- 
tude implies expensive transportation. Within sight of plains 
civilizations are villages of meager and almost medieval culture. 
The mountain masses about the university towns of Grenoble 
hold wretched folk struggling almost with their bare hands 
against the catastrophic forces of nature. The shepherd in the 
Tirolean Alps, sleeping in a hut that is merely a crude pile of 
stones, looks down on the brilliantly lit towns of the Inn Valley. 
Families in the Swiss Alps may winter in the same room as the 
beasts, dependent upon them for warmth, and yet in the darkness 
of the brutish cave hear the shrill whistle of the Simplon Express 
rushing between Paris and Milan. Let us remember that Switzer- 
land is the country of highly cultivated Geneva, Bern, and 
Zurich. Nowhere on the plains is there discoverable so rapid a 
gradient of culture. 



226 MOUNTAIN GEOGRAPHY 

The difficulty of transportation is, of course, a chief factor in 
the retarding of the upward advance of civilization. Literally 
thousands of towns in the mountains of Europe have not yet 
roadways. Twelve villages in Andorra are connected with the 
outside world, the doctor, and educational systems, by a single 
path. Even in the enlightened Swiss Alps people live part of 
the year in alp villages that are not always easily accessible, while 
the ' permanent' villages may be cut off by snows and avalanches 
in the winter. We have told elsewhere how the telephone and 
telegraph are the first means of communication to go out with the 
coming of winter, how the sick lack physicians and the dead are 
unburied. 

The isolation is not merely a question of the difficulty of trans- 
portation. If the resources of the region are sufficiently great, a 
roadway or a railroad is built. Even regions of meager economy, 
as the remote and elevated high villages of the Oisans, are rapidly 
being reached by fine roads through the enlightened policy of 
the national governments. The nations which surround the Alps 
are projecting roads in the face of great difficulties, reaching the 
most remote communities with telegraph and postal systems, 
and sending out school teachers as missionaries. Nevertheless, 
the task is far from complete and the degree of success will always 
be relative. (Figures 64 and 65.) 

Even roads, therefore, are not sufficient to bring in civilization 
if the resources are slight. A study was made by the writer of two 
neighboring villages in the Mediterranean Pyrenees of France. 
One, the village of Py, is connected with the outside world by a 
road. The elevation of the village is not great, 1000 meters, and 
the distance from the railway about 20 kilometers. The settle- 
ment has 1000 inhabitants. The other village is Mantet, reached 
best from Py by the Col de Mantet, 1 765 meters. Py, then, lies in 
a different valley from Mantet. The latter village has no proper 
outside communication along its own valley, because the stream 
has cut so deep a gorge that even trail building is impracti- 
cable. There is no road to Mantet, only the rocky trail shown in 
Figure 63. 

The irrigated hay fields of Py are not continuous with those of 
the lower valley because of a gorge. The village consists of crude 
unplastered houses. There are a well built school and a tiny 
town hall, both of modern construction. All other buildings are 
marked by their lack of mortar and few windows. 




Fig. 63. The Road to Mantet, French Pyrenees 

This is the route by which Mantet maintains communication with the 

outside world. 




Fig. 64. Road Engineering, Switzerland 

l|, (M „ 0< |„n „, il ,h«v.,|, F „;Klim !; ll.., M I i.rt.hanJ V.,s on ,»,<■ h.-Jmn Mi 
(Ooiuirsy <>i' (.iinn ami ( :< .u q >.< n> , .V'. < .„/,..■: (.r^i.ifwt.) 



THE CHARACTER OF MOUNTAIN LIFE 227 

Though the valley bottoms hold the necessary vegetable 
patches, the better lands are given over to irrigated hay fields. 
Grain is grown on terraces high on the mountains where the 
hours of sunlight are longer. The grain is chiefly rye; the people 
eat black bread. A little wheat is produced, and also maize, 
which is cut green for fodder. The scarcity of grain means little 
straw, and the cattle are dirty because of the lack of litter for 
bedding. The highest wheat field discovered was at 1204 me- 
ters, but rye was found at 1 700 meters. In the upper levels are 
straw-thatched barns for the hay and grain harvest. The money 
crop is meat, produced from beasts that are pastured above the 
tree zone from June 1 to October 15. 

Mantet consists of several score houses, built of rough field 
stone. The houses are ill kept and everywhere there are signs of 
decay. Manure piles, of several years' accumulation, lie in the 
irregular streets. The air during the writer's visit was full of dust 
from the dirty streets, and most of the people had diseased eyes. 
A quick survey indicated the majority of the inhabitants to be of 
subnormal mentality. Imbeciles were not lacking. There was a 
school, but little evidence of education. The church was in need 
of repair and without a regular priest. Outside of the church, a 
forgotten symbol, and the school, there was no apparent social or 
economic organization to the village: no store, no inn. One girl 
in the village looked clean and had some bright color to her 
dress. Was she a visitor? The agriculture was the most primitive 
that the writer discovered in the Pyrenees. Here altitude, isola- 
tion, rocky gorge, and rocky trail had completely defeated prog- 
ress in human development. It was difficult to conceive of the 
settlement with its decadent culture as a part of enlightened 
France. 

This brief description of Mantet is an example of low stand- 
ards of living among mountains. It is not typical. Most moun- 
tain villages in Western Europe have the advantages of good 
roads, fine schools, religious leadership, and an active commercial 
connection with the main valley or lowland. The amount of 
electrification in mountain villages is a happy surprise. The Alps, 
be they French, Italian, Swiss, or Austrian, are provided with 
inns serving a fastidious tourist trade. The tourist industry has 
penetrated to many of the more isolated settlements because of 
the search for solitary beauty and for the quaintness which goes 
with isolation. Indeed, one's sense of proportion is strained 



228 MOUNTAIN GEOGRAPHY 

when, as an Alpinist visiting the Alps, one discovers how many 
peaks may be mounted comfortably in a funicular or cable rail- 
way. There is a luxurious hotel near the summit of the Zug- 
spitze. One can dine well in a modern building at the Jung- 
fraujoch overlooking the source of the Aletsch glacier. St. 
Moritz, one of the highest inhabited towns of the Alps, is glitter- 
ing and sophisticated. 

The most important breaking down of mountain provincial- 
isms is along pass routes. The great routes through the Alps are 
modernized to serve the worldly tastes of the travelers. Because 
the history of travel through passes is the story of plainsmen, it 
hardly deserves a place in a volume devoted to mountains and 
mountain folk. Indeed the subject is worthy a volume by itself. 
A brief bibliography of this interesting subject is given at the end 
of the chapter. 

Seasons in the Alps 

The third social trait of mountaineers is, in the case of truly 
alpine mountains, industry in conflict with almost overwhelming 
toil. Part of the excessive labor is a consequence of the meager 
resources, but some of it arises from the nomadic movements, in 
part pastoral and in part agricultural, which carry the weary 
population up and down the slopes. We can best approach a 
conception of labor in mountains by considering seasonal work 
in the Alps. We do this in spite of the fact that one has little 
right to make a general description of life in so long and varied a 
mountain range. Many purely local cultures are found there, 
differentiated by topography, climate, and custom. 

At the southern end, in the Alpes-Maritimes of Provence and 
Nice, we have the home of the goatherd. The town of his origin 
is a fortified village, perched on the rocks, still with legends of 
Saracen raids. With his goats the herder spends the day among 
the thyme moorlands, returning at night to the village, where 
each beast is milked in the stable of the owner. At the Tirolean 
end of the Alps there is an idyllic picture, now all but gone. This 
is of the girl who watches the cattle on a lonely but serene alp- 
land, above the village and above the tree line. For weeks she is 
alone with her beasts, the flowers of the meadows, and the circle 
of clear-cut peaks, alone except for Sunday visits of her lover. 

For a typical dairy scene let us return to a more central loca- 



THE CHARACTER OF MOUNTAIN LIFE 229 

tion, Switzerland. The scene, a busy one, is of some five or more 
men engaged with the cattle. The leader of the group is a cheese 
maker, upon whose skill much of the reputation of the valley 
rests. He will be assisted by several herders and milkers. There 
is one man whose task it is to supply wood for the boiling of the 
milk, and another worker may care for the alp itself, clearing it of 
obnoxious weeds and building barriers in the gullies to prevent 
erosion. If there is a wagon or sled road leading to the valley, the 
animals may be corraled at night and the manure thus collected 
for transportation to the fields. In any case, by trail or road, sup- 
plies must be taken up to the men, and the cheeses carried valley- 
ward. There is a variety of wood or wicker burden frames which 
are attached to the backs of the men who perform this transport 
duty. 

The alp pastures are reached in early spring. Here it is that 
the herdsmen's life begins in earnest. A cluster of small stone 
huts marks the first stage of the nomadism. The most commo- 
dious one is selected by the chief herdsman and his helpers. The 
others are shelters for the cattle and perhaps a special hut for the 
calving cow. The chief hut usually has one living room and a 
milk room. At one side of the living quarters hay is spread upon 
the floor for a couch. The herdsmen arise shortly after dawn. 
As the distant church bells ring, they say a short morning prayer 
upon their rosaries. The cows are allowed to graze in the early 
morning while the herdsmen proceed to the cheese making. The 
best cheese is made from milk fresh from the cow. 

The cattle may be taken back into the stables when the heat 
from the midday sun becomes too strong. The herdsmen eat 
their dinner and lie down for a rest. Towards evening, when the 
mountain air is again cool, the cattle are released, the wood is 
collected, the water for the next day is brought, evening prayers 
are said, and thus another day has passed. 

In the meantime the rest of the population is busy about the 
labor of the valley farm. Mountain labor is hard, and mountain 
livings are meager. There is little level land, and soil may be 
thin on slopes. A score of different catastrophes may overtake 
the crop and even the field. The mountain earth itself, as a land- 
slip or a torrent, may overwhelm the crop, road, or village. 
Mountain weather in its extremes is destructive. Peasants have 
been known to keep the grain of a successful harvest for four or 
five years for fear of starvation. The higher the village the more 



230 MOUNTAIN GEOGRAPHY 

work necessary to maintain life, for communications are main- 
tained with difficulty, travel is arduous, freightage is expensive, 
the winter period is longer, the fields small, and on precarious 
slopes catastrophes are frequent. It is the burden of labor in the 
mountain village which explains why so many avail themselves 
of the first opportunity of emigration to the plains. 

The communication trails between the high regions and the 
valley village are by their necessary and frequent repairs a con- 
siderable burden upon the commune. But the true labor of the 
trails is the frequent passage, the back-breaking trips up the 
mountain and the tedious descents. The high field may be one 
and two hours' climb from the village. Up this slope, men, 
women, and children toil with deep manure baskets on their 
backs. Down the trail the rye is carried in sheaves upon the 
head (Figures 62). The potato fields of the Spanish Sierra 
Nevada are three and four hours up the trail from the villages 
(Figure 61), and the chestnut zone an hour's climb. Peasants in 
the Pyrenees climb three hours to reach alps (Jasses) where they 
find a plant good for salads. The weekly journey of a purveyor 
from village to summer alp is not infrequently an all-day trip. 

Perhaps no form of agriculture is more laborious than terrace 
tillage (Figure 49). The building of the terraces and their repair, 
the control of rainwater and irrigation, the fertilizing of the fields 
and the harvest, call for labor which may leave the peasant with 
a damaged heart and certainly a bent back. Terraces lose soil by 
washing. This dirt must be carried back 'up the steps.' 

The higher the village the more desperate the agriculture, and 
the greater the attention and area given to hay. In these high 
altitudes crops will not mature. This is true especially on shady 
slopes. But in many easily accessible valleys there may be, and 
often are, areas so isolated as not only to be prohibitive to tillage, 
but impossible to be reached by beasts without travel on dan- 
gerous trails. Thus the village of Saint-Martin-de-la-Porte 
(Maurienne), at 820 meters, has hay meadows in the Col des 
Encombres at 2300 meters. The difference in elevation is 1680 
meters or 5600 feet. Down steep mountain trails perhaps but one 
load of 100 kilograms represents a day's work, and it is good for- 
tune if one is not caught on the way by a thunder-shower. Be- 
tween Ch&teauroux in Embrunais and certain hay fields mules 
will make two trips a day, bringing 50 to 60 kilograms a trip. 
The day starts at midnight and ends at 8 in the evening. Many 



H 

■■ill 

111 


mm 


"Iff 



■J3BBM : 



» #. & ""w. j .'-?? , --^? j *-,-.-i 

p WHt » ShR <- HI 




S.S 






(X 2 



'21 



£c 3 

G fa ,*"" 



•S-M 



I— i c > ?? 



■8.8 3 

jy -a £* 



THE CHARACTER OF MOUNTAIN LIFE 231 

hay fields are perched so dangerously that there are communal 
regulations to guard the lives of hay gatherers. 

The bringing of the hay to the valley stable presents a problem 
(Figure 62). Remarkable loads are carried down the slopes on 
frames set on men's backs. Another worker helps the burden 
bearer stagger to his feet with the excessive load. The bearer 
then clambers down a half hour's descent to the village. A horse 
may bear a load on a frame, the hay all but hiding the beast. 
Aerial cables are used in many places. Some in the valley of 
Naves are described in an earlier chapter. In the villages above 
Chur, in the Grisons, sleds bring hay in autumn down slopes im- 
passable for carts. In the Val d'Anniviers, as elsewhere, sleds 
bring hay down in winter. Haying occupies most of the summer. 

In September the herdsmen begin their descent from the alp 
pastures with their herds. The people in the valleys ascend the 
mountains to the village on the may en alp at the same time. 
When the entire village is again united there is a joyful celebra- 
tion marked by dancing, drinking, and feasting. A hay crop is 
harvested from fields about the 'temporary' village, as are also 
the vegetables that were planted when the alp-drive started in 
May. The work here completed, the entire group move down- 
ward to spend the winter in the valley. 

With the coming of the heavy snows, the cattle are put in the 
barns and not taken out again until spring. The cold is so in- 
tense that the houses and barns are almost hermetically sealed. 
Because of this enforced imprisonment, fuel and food must be 
stored in sufficient quantities in the autumn to last through the 
winter. It is this period of confinement that creates the lace- 
making, wood-carving, watch-making industries of some parts 
of the Alps. Not everywhere, however, is the confinement of the 
same degree. In the Fore-Alps winter does not preclude out-of- 
door work. There is hay to be transported, manure to be spread 
upon the fields, and wood to cut. The women will wash clothes 
by a hole in the ice at sub-zero temperatures. The higher the 
village the longer the duration of the snow. In turn, then, the 
longer the period of stable feeding and the greater the area of 
mown grass. The higher the village, the greater the number of 
hay barns, and the steeper the pitch of the roof in order to be rid 
of snow. 

The confinement of winter is oppressive in a degree to be meas- 
ured by the joy expressed by the spring liberation. The high 



232 MOUNTAIN GEOGRAPHY 

insolation and the warmth, once the snow has left and no longer 
chills the air, are particularly delightful. The valley town may 
be in the shadow of the mountain so that the south-facing alp 
first loses its snow cover. Eager flowers literally push up through 
the snow. Schools are dismissed so that the children may play 
among the spring flowers. The first influx of visitors begins, 
coming to look upon the fields of narcissus. Soon the surge of 
life is felt throughout the valley. The mountainward migration 
is about to begin. Though today much of the picturesqueness 
has given way to the modern spirit, the customs of the past are 
still found in certain isolated villages. The herds are assembled 
in a large procession, each one preceded by a herdsman and a 
flock of goats. The herdsman may wear a white shirt with 
sleeves rolled to the elbows, a pair of broad leather suspenders 
decorated with figures of cows and goats shaped from bright 
metal, a scarlet waistcoat, knee trousers of yellow cloth, white 
stockings, low shoes, and a round black hat of felt or leather 
bound with a wreath of artificial roses. The main body of cattle 
is kept in line by a herdboy. Even the cows, wearing wreaths, are 
eager to be off. Each file of herds may be followed by a wagon or 
pack train, containing a great copper cheese-kettle and wooden 
utensils for milk and butter. The procession moves along the 
road while great bells clang. At the entrance to every inn, the 
landlord issues forth with decanter of wine and glasses to serve 
the herdsmen. Many times the majority of the population of the 
valley villages accompany the herdsmen and herds to the middle 
pastures or may ens. The first hay crop is gathered. Vegetables 
are planted in well manured fields. The children play in the 
warm sunshine. There is much activity; everybody is busy. 
Soon the cows begin to mount to higher pastures and the sheep 
leave for the very highest areas. The rhythm of life recommences. 

The Tenor of Andorran Life 

Lastly, by way of conclusion, let us consider Andorra as a 
mountain province, illustrating at once conservatism, low stand- 
ards of living, and hard labor. Withal we hope to reveal a sim- 
plicity of life and a beauty that many a poet has felt the lack of 
words to express. 

Andorra is alternate savage gorge and open valley. It has 
nothing of the complacency of the gaves of the Central Pyrenees. 



THE CHARACTER OF MOUNTAIN LIFE 233 

Its mountains are gaunt things. Talus and glacial chaos im- 
pend above the fields and threaten the villages. Nature here is 
catastrophic. It is a land of sheer cliffs that cast deep blue 
shadows over cold canyons, while far above, amid pink and 
purple mountains veiled with the green of conifers, are forested 
valleys. Here echo cascades of limpid water, ideal haunts for the 
fisherman. Yet higher are rocky cirques with deep blue lakes, or 
the dreary moors that pasture sheep. 

It is a dramatic little land, romantic and tragic. One feels it 
in the solemnity of the native. It is a land of toil. The casual 
traveler sees only the fertile valleys. But high on the mountains 
are pitiful patches of rye which give but meager returns. Even in 
the valley bottoms there is scarcely enough sun to cure the hay. 
Above the zone of desperate agriculture are the desolate pas- 
tures, where the life of the shepherd is indescribably lonely. 

Villages such as Sant Julia, Andorra la Vieja, and Escaldes are 
in fine valleys, on a highway, and have a brisk trade. But visit 
Pal, hours from the road; Arinsal, at the foot of a mountain 
cirque; and Lo Serrat, most distant of Andorran hamlets. On 
the plea of thirst I entered one home in Arinsal. The way in was 
through the pig sty, for the pigs held the ground floor. Above, the 
long living room was blackened by the pine torches used for 
lighting. Tiny windows let in such light as the darkling sky gave 
forth. The floor was littered with bean vines that were to be 
picked over in the evening. The baby in the cradle cried because 
of the flies that crawled over its face. The pigs could be seen, 
heard, and smelled through the wide cracks of the floor. Out- 
side, the stream and canyon wind roared in competition. 

But Andorra has likewise its charm. Every turn of the road 
brings scenes that delight the traveler. The first things that made 
me feel that I truly had found a national individuality charac- 
teristic of the region were the old bridges. Slender, long, grace- 
ful arches made of cobbles or split stone, half their charm lies in 
their lack of symmetry and in their imperfection. The villages of 
dark-brown cobbles are more charming for the long green and 
brown streamers of drying tobacco leaves that hang from the 
windows or that fill the open shed on the house-top. The An- 
dorran loves his balcony, and whether with wooden railing or 
simple iron work, it is a delight. Commonly the street between 
the houses is so narrow that a single horse, laden with brushwood 
from the mountains or with hay from the fields, entirely blocks 



234 MOUNTAIN GEOGRAPHY 

the passage. The tobacco comes to the shed suspended in ham- 
mocks on the sides of a horse, or its green leaves fill a two-wheeled 
cart drawn by oxen. 

Let us attend a fete in the village of Ordino. Picture a village 
square, the church, the stone steps, the little alleyways leading off 
between the houses, the garlanded cross, the flowing fountain, 
and every window and balcony filled with onlookers. On a plat- 
form are the six Spanish musicians with horns. All about are the 
cliffs and dark forest, and above, a deep blue sky. As the music 
starts each girl walks to the stone bench, where, sitting in a row, 
the belles are to wait till the swain approaches and nonchalantly 
beckons one of them. The dance is quite a la mode except that 
there is a leaning towards the polka. Alas, the days have passed 
when a gallant wore knee-breeches and a scarlet Jache and beret, 
and the lady had a lace head-dress, a low bodice, and a flowing 
skirt short enough to show her fine petticoat. Now the dance is 
modern enough, though done quite solemnly. 

Let me paint as best I can an evening in a little fonda far from 
the highway. 1 Picture yourself sitting before a hearth which is 
overhung by its great mantel, blackened with smoke. On a crane 
over the fire hangs the pot in which our supper, a rabbit, is boil- 
ing. A man tired with the mountain trail is sleeping, his head in 
his arms upon the table. Other men drop in for a bit of cheer. 
Outside, the mountain reflects the last rays of the sun. Finally 
darkness. Within we have our fire and the light reflected from 
the copper pots that hang in a row on the wall. 

From a vantage on a mountain spur the canyon-like valley 
could be surveyed. 1 saw a bent man following the plow slowly 
dragged by oxen. The field in the valley, chilled by drainage of 
cold air and refrigerated by an icy stream, required thirteen 
months to produce a rye crop. This was an October evening and 
the field was being plowed for an autumnal sowing. The shadow 
of the mountain came down over field and farmer. The valley 
became deep purple in color, and cold. In the dim light, the 
man, crippled by labor in these mountains which impended 
above him, followed the plow. Valleys literally hide their in- 
habitants. Mountain shadows chill the hearth stone. 

I think rny favorite spot in Andorra is in the village of Engor- 
dany. The houses of the village cluster on a hillside at the end of 
a valley. And though I know and love the spot at all hours of the 

1 The reader will kindly permit the first person for the sake of directness. 



THE CHARACTER OF MOUNTAIN LIFE 235 

day, it is at dusk that I love it best. Then the rush of the waters 
in the stream seems most lovely. Then the valley lands are 
deepest green. The jagged Sierra d'Enclar is a deep purple 
against a fading sky. One single light shows the location of An- 
dorra la Vieja at the foot of its mountain. Far down the valley, 
just over the Sierra de Leix, rises a star. The peasants pass before 
me, saluting me with Catalan salutations or with the deep-toned 
Buenas tardes. Here comes the bulk of a donkey all but buried in 
the hay it carries, the haying party shuffling along behind. From 
the other direction comes a little herd of sheep following the 
shepherd, who whistles encouragingly to them. Then the priest 
passes. A girl comes along with her pigs from their day's grub- 
bing on the mountain. All of the little world passes before me in 
the growing darkness. Twilight has gone. Night has arrived. 
In the heavens the lesser stars are come forth. One then takes the 
road that passes the stone bridge to the cheer of the inn. 

Man has modified the face of the mountains by stupendous 
labor, but the great earth masses that all but blot out the stars of 
night, and by day limit the sunshine which may penetrate to the 
fields in the canyons, condition, and always will condition, man's 
activities. 



BIBLIOGRAPHICAL NOTES 

The Appalachian Mountaineers 

Brigham, A. P. "The Appalachian Valley," in Scottish Geographical 

Magazine, xl (1924), pp. 218-230. 
Campbell, J. C. The Southern Highlander and his Homeland. New York, 

1921. 
Davis, D. H. " The Changing Role of the Kentucky Mountains and the 

Passing of the Kentucky Mountaineer," in Journal of Geography, 

xxiv (1925), pp. 41-52. 
Davis, D. H. The Geography of the Mountains of Eastern Kentucky. Frank- 
fort, Ky., 1924. (Kentucky Geological Survey, Geologic Reports, 

series vi, xviii.) 
Davis, D. H. " A Study of the Succession of Human Activities in the 

Kentucky Mountains," in Journal of Geography, xxix (1930), pp. 85- 

100. 
Hollander, A. N. J. den. " Uber die Bevolkerung der Appalachen," in 

Z^itschrift der Gesellschaft fiir Erdkunde zu Berlin, 1934, Heft 7-8, pp. 

241-255. With a bibliography. 



236 MOUNTAIN GEOGRAPHY 

Kephart, Horace. Our Southern Highlanders. New York, 191 3. An 
authority. 

Mason, R. L. The Lure of the Great Smokies. Boston, 1927. 

McClarty, Julia. " Economic Opportunities in the Southern Appala- 
chians," in Journal of Geography, xx (1921), pp. 96-104. 

Morley, Margaret Warner. The Carolina Mountains. Boston, 191 3. 

Schockel, B. H. " Changing Conditions in the Kentucky Mountains," 
in Scientific Monthly, iii (191 6), pp. 1 05-131. 

Semple, Ellen Churchill. "The Anglo-Saxons of the Kentucky Moun- 
tains," in Bulletin of the American Geographical Society, xlii (1910), pp. 

5 6l ~594- 
Spaulding, A. W. The Men of the Mountains. Nashville, 1915. 

Mountain Passes: Together with Characteristic References on 
Mountain Railway Construction 

The subject of mountain passes is an interesting one which is con- 
nected with that of mountain isolation. However, passes are more im- 
portant to extramontane areas than to the mountain regions which they 
traverse. The Brenner, St. Gotthard, and Mt. Cenis passes have been 
more significant to Venice, Milan, and Genoa, to Lyons, Augsburg, and 
Innsbruck, than to the mountain regions. Only occasionally has a 
mountain city like Chur (Coire) been greatly influenced by travel 
across the passes. In this volume, which is devoted to the mountain 
areas rather than to the influence of mountains on remote areas, the 
subject can be given little place. A bibliography is here given for those 
who wish to pursue the subject. 

Allix, Andre. "Le trafic en Dauphine" a la fin du Moyen-Age," in 

Revue de geographie alpine, xi (1923), pp. 373—420. 
Arbos, Philippe. "Les communications dans les Alpes francaises," in 

Annates de geographie, xxviii (191 9), pp. 1 61-176. 
Blanchard, Marcel. Les routes des Alpes occidentals a Vepoque napoleonienne 

( 1 796-1815). Grenoble, 1920. 
Blanchard, Raoul. " Le r£seau ferre des Alpes franchises," in Recueil des 

travaux offert a M. Jovan Cvijic (Belgrade, 1924), pp. 233-240. A 

good example of mountain railway geography. 
Blanchard, Raoul. " Les zones d 5 6quidistance des voies ferr6es dans les 

Alpes franchises," in Revue de geographie alpine, xii (1924), pp. 79-97. 
Bonney, T. G. The Alpine Regions of Switzerland and the Neighbouring 

Countries. Cambridge, England, 1868. 
Bonney, T. G. "The Alps from 1856 to 1865," in Alpine Journal, xxxi 

09i7)> PP- l6 ~34. 
Chabot, Georges. "Les perc6es des Vosges," in Annales de geographie, 

xxix (1920), pp. 376-378. 
Dainelli, Giotto. " Italia Pass in the Eastern Karakoram," in Geographic 



THE CHARACTER OF MOUNTAIN LIFE 237 

cal Review, xxii (1932), pp. 392-402. No attention in this present 

volume has been paid to Alpinism. Here is a splendid example 

which at once describes a high pass and the rigors of mountain 

climbing. 
Featherstone, B. K. An Unexplored Pass: A Narrative of a Thousand- Mile 

Journey to the Kara-Koram Himalayas. London, 1926. Material on 

heights of habitation and culture of Tibet. 
Fliickiger, Otto. " Passe und Grenzen," in Mitteilungen der Geogra- 

phisch-Ethnographischen Gesellschqft in Zurich, xxvii-xxviii (1926-28), 

pp. 36-65. Very important. 
Freshfield, D. W. "The Great Passes of the Western and Central 

Alps," in Geographical Journal, xlix (191 7), pp. 2-26. 
George, H. B. The Relations of Geography and History, 4th ed. Oxford, 

1 910. Chap, xiv, "The Alpine Passes and their History." 
Girardin, Paul. "Les bassins fermes des Alpes suisses," in Bulletin de la 

Societe Fribourgeoise des sciences naturelles, xxii (19 14), p. 15. 
Johnson, D. W. "How Rivers cut Gateways through Mountains," in 

Scientific Monthly, xxxviii (1934), pp. 129-135. 
Langwill, Minnie J. " Historic Mountain-Passes of the World," in Jour- 

nal of Geography, xii (1913-14), pp. 193-197. 
Montzka, H. "Die Mittenwaldbahn," in Deutsche Rundschau fur Geo- 

graphie, xxxiv (191 2), pp. 428-434. An example of the difficulty of 

railway building in mountains. 
Onde,'H. " La route de Maurienne et du Cenis de la fin du XVIII e au 

milieu du XIX e siecle," in Revue de geographie alpine, xx (1932), pp. 

701-775. 
Palmer, A. H. "Snow and Railway Transportation," in Monthly 

Weather Review, xlvii (191 7), pp. 698-699. 
Preller, C. D. R. "Hannibal's Passage of the Alps," in Scottish Geo- 

graphical Magazine, xlii (1926), pp. 350-359. 
Sargent, A. J. "Alpine Railways and International Commerce," in 

Geographical Journal, xxv (1905), pp. 654-658. 
SchefTel, P. H. Verkehrsgeschichte der Alpen. Berlin, 1908-14. 2 vols. A 

history of Alpine commerce. There are numerous books and articles 

on this subject. 
Semple, Ellen Churchill. Influences of Geographic Environment. New 

York, 191 1. Chap. xv. 
Semple, Ellen Churchill. "Mountain Passes: A Study in Anthro- 

pogeography," in Bulletin of the American Geographical Society, xxxiii 

(1901), pp. 124-137, 191-203. 
Solch, Johann. Studien iiber Gebirgspdsse. Stuttgart, 1908. (Forschungen 

zur deutscher Landes- und Volkskunde, xvii, 2 .) The authoritative Ger- 
man work. 
Tyler, J. E. The Alpine Passes: The Middle Ages (962-1250). Oxford, 

1930. An excellent history. Especially recommended. 



238 MOUNTAIN GEOGRAPHY 

Ver Steeg, Karl. " Wind Gaps and Water Gaps of the Northern Appa- 
lachians," in Annals of the New Tork Academy of Sciences, xxxii (1930), 
pp. 87-220. 

Some References on the Quality of Mountain Life 

Baillie-Grohman, W. A. Tyrol and the Tyrolese. Leipzig, 1877. 
Baud-Bovy, Daniel. Peasant Art in Switzerland, tr. by Arthur Palliser. 

London, 1924. A magnificent book. 
Belloc, Hilaire. The Pyrenees. London, 1909. A pleasing writer. 
Coulter, J. W. "Lucia: An Isolated Mountain District in California," 

in Bulletin of the Geographical Society of Philadelphia y xxix (1931), 

pp. 183-198. An example of surprising isolation in America. 
Cowan, A. R. Master-Clues in World-History. London, 19 14. Chap. 

viii. 
Demarez, R. "Les modes de vie dans les montagnes de l'lndo-Chine 

franchise," in Recueil des travaux de ITnstitut de geographie alpine, vii 

(1919), PP- 453-561. 
Hilton-Simpson, M. W. "The Influence of its Geography on the 

People of the Aures Massif, Algeria," in Geographical Journal, lix 

(1922), pp. 19-36. 
Lane, Rose Wilder. The Peaks of Shala. London, 1922. Isolation in 

Albania. 
Legendre, Maurice. Las Jurdes. Bordeaux, etc., 1927. {Bibliotheque de 

r Ecole des hautes etudes hispaniques, fasc. xiii.) Reviewed by Raoul 

Blanchard in Revue de geographie alpine, xvi (1928), pp. 545-551. A 

primitive culture. 
Martonne, Emmanuel de. " Quelques donnees nouvelles sur la jeunesse 

du relief preglaciaire dans les Alpes," in Recueil de travaux offer t a 

M. Jovan Cvijic (Belgrade, 1924), pp. 1 21-140. Includes physical 

isolation of valleys of the Var drainage system. 
Meylan, Rene. " La vall6e de Joux: Les conditions de vie dans un haut 

bassin ferme du Jura," in Bulletin de la Societe neuchdteloise de geo- 
graphie, xxxviii (1929), pp. 45-179- Social isolation. 
Miller, Mrs. Anna C. (Johnson). The Cottages of the Alps. New York, 

i860. Good. 
Oakley, Amy. Hill-Towns of the Pyrenees. New York, 1923. 
Peattie, Roderick. New College Geography. Boston, 1932. Chaps, xxi 

and xxii. 
Peattie, Roderick. " Wanderungen in Andorra," in Der Erdball, iv 

(i93°)> PP- 287-290. 
Peattie, Roderick. " Catalan France and the Canigou," in Ohio Social 

Science Journal, i (1929), pp. 36-43. 
Rawnsley, H. D. Flower-time in the Oberland. Glasgow, 1904. 
Rey, Maurice. "La limite geographique de l'habitat perch6 dans les 



THE CHARACTER OF MOUNTAIN LIFE 239 

Alpes frangaises," in Revue de geographie alpine, xvii (1929), pp. 5-39. 
Relationship between nomadism and the village type. 

Robert, Jean. " Un type de cohabitation avec les animaux," in Revue 
de geographie alpine, xxi (1933), pp. 819-829. Primitive pastoral life. 

Rock, J. F. "Seeking the Mountains of Mystery," in National Geo- 
graphic Magazine, lvii (1930), pp. 1 31-185. Remoteness in China. 

Semple, Ellen Churchill. Influences of Geographic Environment. New 
York, 191 1. 

Slovene Studies, by members of the Le Play Society, ed. by L. D. Stamp. 
London, 1933. Excellent. 

Solch, J. "Raum und Gesellschaft in den Alpen," in Geographische 
Zeitschrift, xxxvii (1931), pp. 143-168. Philosophical. 

Stephen, Leslie. The Playground of Europe. London, 1871; new ed., 
1894. A famous book on Switzerland. 

Story, A. T. Swiss Life in Town and Country. London, 1902. Informa- 
tive. 

Tissot, Victor. Unknown Switzerland, tr. by Mrs. Wilson. New York, 
1889. 

Webb, Frank. Switzerland of the Swiss. London, 1909. 

Wilstach, Paul. Along the Pyrenees. Indianapolis, 1925. 



APPENDIX A 

THE METRIC SYSTEM 

WE are not so much inclined to apologize for using the metric 
system as to regret that we as Americans are not more thoroughly 
acquainted with it. As most of the material of this study, of an exact 
sort, comes from French, German, and Italian sources and is in the 
metric system, it is important that the metric system be here used. To 
convert these figures into the systems used in America would be to risk 
accuracy. In some cases, on the other hand, data of the English system 
are converted to the metric. Where this has not been done, the reasons 
will be obvious. The conversion of feet to meters has been accomplished 
by the multiplier 0.305. Miles have been multiplied by 1.6 in the con- 
version to kilometers. Simple rules for converting meters to feet and 
kilometers to miles are given. If these are memorized the reader can 
convert to the English system subconsciously while reading. 

For Heights 

If meters are divided by 3 and the decimal shifted one point to the 
right, the height in feet is obtained within a certain limit of error. The 
exact constant of the error is 1 .6 feet too much for every 100 feet. Thus, 
if an elevation stands at 3048 meters, one divides by 3 and adds a zero, 
giving 10,160 feet. After applying the correct elevation, it is repre- 
sented as 10,000 feet. It is not always easy to compute the amount of 
error. This may be put in a formula: cut off the last three figures, mul- 
tiply by four, multiply again by four, and subtract from the first rough 
calculation. 

Thus an elevation stands at 2403 meters. 

3 I 2403 8 X 4 is 32 32 X 4 is 128 8010 

" 801 (o) 128 

7882 

This is within 20 inches of the truth. The exact height of the eleva- 
tion in English feet is 7883.7624 feet. This is obtained by multiplying 
2403 by 3.2808992. 

For Distances 

One divides kilometers by 8 and multiplies by 5. The result would be 
correct if the kilometer were equal to 3300 feet. The fact that in reality 
it is approximately 3281 feet indicates the limits of error. 



244 APPENDIX A 

For Climatic Data 

In converting pressure and precipitation data the following formulas 
are sufficiently accurate: 

10 mm. = 0.394 inches 
1 inch = 25.40 mm. 

The conversion formula for Centigrade (C) to Fahrenheit (F) is 
C = F - 32 

5 9 ' 

In practice the following table has been used: 

Conversion Table for Temperature 



°G 


°F 


°G 


°F 


— 22 


— 7.60 


6 


42.80 


— 21 


-5.80 


7 


44.60 


— 20 


— 4.OO 


8 


46.40 


-*9 


— 2.20 


9 


48.20 


-18 


— O.4O 


10 


5O.OO 


-17 


i .40 Plus 


1 1 


51.80 


-16 


3.20 


12 


53-60 


-15 


5.00 


13 


55-40 


-14 


6.80 


M 


57.20 


-13 


8.60 


*5 


59.00 


— 12 


10.40 


16 


60.80 


— I I 


12.20 


i7 


62.60 


— IO 


14.00 


18 


64.40 


~ 9 


15.80 


*9 


66.20 


- 8 


17.60 


20 


68.00 


- 7 


19.40 


21 


69.80 


- 6 


21.20 


22 


71.60 


- 5 


23.00 


23 


73-40 


- 4 


24.80 


24 


75.20 


- 3 


26.60 


2 5 


77.00 


— 2 


28.40 


26 


78.80 


— 1 


30.20 


27 


80.60 


— 


32.00 


28 


82.40 


+ 1 


33.80 


29 


84.20 


2 


35-60 


30 


86.00 


3 


37-40 


3i 


87.80 


4 


39-20 


32 


89.60 


5 


41.00 


33 


91.40 



APPENDIX B 

BIBLIOGRAPHICAL NOTES ON THE GEOMORPHOLOGY 
OF MOUNTAINS 

The Origin of Mountains 

The treatises named below are selected as characteristic of this field 
of knowledge. Some are reviews, while others are recent contributions 
to theory. 

Bonney, T. G. The Building of the Alps. London, 191 2. 

Bonney, T. G. Volcanoes: Their Structure and Significance. London, 

1899. 
Bowie, William. Jsostasy. New York, 1927. An authority. 
Bowie, William. " Proposed Theory in Harmony with Isostasy to Ac- 
count for Major Changes in the Elevation of the Earth's Crust," in 

Beitrage zur Geophysik, xv (1926), pp. 103-115. 
Chamberlin, R. T. "The Building of the Colorado Rockies," in Jour- 

-nal of Geology, xxvii (1919), pp. 145-164, 225-251. 
Chamberlin, R. T. "On the Crustal Shortening of the Colorado 

Rockies," in American Journal of Science, 5th ser., vi (1923), pp. 215- 

221. 
Chamberlin, R. T. " Vulcanism and Mountain-Making," in Journal of 

Geology, xxix (192 1), pp. 166-172. 
Collet, L. W. The Structure of the Alps. London, 1927. 
Daly, R. A. Our Mobile Earth. New York, 1926. Excellent. 
Dana, J. D. "On the Origin of Mountains," in American Journal of 

Science, 3d ser., v (1873), pp. 347~35°- 
Geikie, James. "The Architecture and Origin of the Alps," in Scottish 

Geographical Magazine, xxvii (191 1), pp. 393-417. 
Geikie, James. Mountains: Their Origin, Growth, and Decay. Edinburgh, 

I9i3- 
Hobbs, W. H. Earth Features and their Meaning. 2d ed. New York, 1931 . 

Chap. xxxi. 
Joly, John. The Surf ace-History of the Earth. Oxford, 1925. Chap. vii. 
Kober, Leopold. Der Bau der Erde. Berlin, 192 1. 
Nevin, Charles M. Principles of Structural Geology. New York, 1931. 

Chap, xi; see also index. Recent and unusually clear. 
Reid, H. F. " Isostasy and Mountain Ranges," in Bulletin of the American 

Geographical Society, xliv (191 2), pp. 354-360. 



246 APPENDIX B 

Salisbury, R. D. Physiography, 3d ed. New York, 191 9. Pp. 377-388. 

A genetic classification. 
Shepard, F. P. " Isostasy as a Result of Earth Shrinkage," in Journal of 

Geology, xxxi (1923), pp. 208-216. 
Staub, Rudolf. Der Bewegungsmechanismus de Erde. Berlin, 1928. Late 

German philosophy. 
Suess, Eduard. Das Antlitz der Erde. Prague, etc., 1 883-1 904. 3 vols. 

English translation by Hertha B. C. Sollas, The Face of the Earth. 

Oxford, 1904-09, 4 vols. The best review of the mountain systems 

of the earth. 
Swanson, C. O. " Isostasy and Mountain Building," in Journal of 

Geology, xxxvi (1928), pp. 411-433. 

Relief Energy 

Relief energy represents a new point of view. It deserves further 
study. 

Glock, W. S. "Available Relief as a Factor of Control in the Profile of 
a Land Form," in Journal of Geology, xl (1932), pp. 74-83. Some 
basic considerations. 

Johnson, D. W. " Available Relief and Texture of Topography," in 
Journal of Geology, xli (1933), pp. 293-305. A reply to Glock. 

Krebs, Norbert. "Eine Karte der Reliefenergie : Siiddeutschlands," in 
Petermanns Mitteilungen, lxviii (1922), pp. 49-53, with map. 

Krebs, Norbert. Die Ostalpen und das heutige Osterreich. Stuttgart, 1928. 
2 vols. Vol. i, pp. 44-46, chart opposite p. 40. 

Ochocka, Janina. "Krajobruz Polski e Sivretle mapy Wysokosci 
Wzgleclnych " (Map of Relative Relief of Poland), in Prace Geo- 
grqficzne wydawane Przez Eugenjusza Romera, zeszyt xiii (1925). A 
geographical treatise under the direction of Eugene Romer. A 
summary in French accompanies the article. It has the most exact 
form of the cartography of relief. 

Partsch, Josef. Schlesien, eine Landeskunde filr das deutsche Volk. Breslau, 
1896-1911. 2 vols. Vol. ii, chart facing p. 586. 

Rich, J. L. "A Graphical Method of Determining the Average In- 
clination of a Land Surface from a Contour Map," in Transactions of 
the Illinois State Academy of Science, ix (1916), pp. 195—199- 

Schrepfer, Hans, and Kallner, Horst. "Die maximale Reliefenergie 
Westdeutschlands," in Petermanns Mitteilungen, lxxvi (1930), pp. 
225-227. Has three references not in this list. 

Smith, G. H. "The Relative Relief of Ohio," in Geographical Review, 
xxv (i935)» PP- 272-284. 

Wentworth, C. K. "A Simplified Method of Determining the Average 
Slope of Land Surfaces," in American Journal of Science, 5th ser., xx 
093°)>PP- 184-194. 



APPENDIX B 247 

The Shapes of Alpine Valleys 

With two exceptions the references are in English. They are of the 
general type. There is abundant material on the subject in French, 
German, and Italian. 

Avebury, Sir John Lubbock, 1st Baron. The Scenery of Switzerland, 

2d ed. London, 1896. 
Davis, W. M. " The Sculpture of Mountains by Glaciers," in Scottish 

Geographical Magazine, xxii (1906), pp. 76-89. A classic. 
Davis, W. M. Geographical Essays. Boston, 1909. Chap, xxiv, " Glacial 

Erosion in France, Switzerland, and Norway." 
Davis, W. M. Die erklarende Beschreibung der Landformen (Leipzig, 19 12), 

pp. 401-462, "Der Glaziale Zyklus." 
Fenneman, N. M. "Some Anthropo-Geographic Effects of Glacial 

Erosion in the Alps," in Journal of Geography, vii (1907-08), pp. 169- 

172. 
Gastaldi, B. "On the Effects of Glacier-Erosion in Alpine Valleys," in 

Quarterly Journal of the Geological Society, xxix (1873), pp. 396-401. 
Gilbert, G. K. "Systematic Asymmetry of Crest Lines in the High 

Sierra of California," in Journal of Geology, xii (1904), pp. 579-588. 
Hobbs, W. H. "The Cycle of Mountain Glaciation," in Geographical 

Journal, xxxv (1910), pp. 268-284. 
Hobbs, W. H. "Studies of the Cycle of Glaciation," in Journal of 

Geology, xxix (1921), pp. 370-386. 
Johnson, D. W. "Hanging Valleys of the Yosemite," in Bulletin of the 

American Geographical Society, xliii (191 1), pp. 826-837, 890-903. 
Johnson, G. R. Peru from the Air. New York, 1930. (American Geo- 
graphical Society, Special Publication no. 12.) 
Johnson, W. D. "The Profile of Maturity in Alpine Glacial Erosion," 

in Journal of Geology, xii (1904), pp. 569-578. 
Johnson, W. D. "The Work of Glaciers in High Mountains," in 

Science, n. s., ix (1899), pp. 112-1 13. An abstract. 
Matthes, F. E. Geologic History of the Yosemite Valley. Washington, 1930. 

Relative importance of glacial and stream erosion. 
Mittelholzer, Walter. Switzerland from the Air, ed. by Otto Fliickiger, 

trans, by A. W. P. Allan. Zurich, 1926, 
Nussbaum, Fritz. Die Taler der Schweizeralpen. Bern, 1910. Included 

because of its general interest. 

The Classification of Mountains 

Bowman, Isaiah. Forest Physiography. New York, 191 4. 
Cleland, H. F. Geology, Physical and Historical. New York, 191 6. Pp. 
352-358. 



248 APPENDIX B 

Emerson, F. V. Manual of Physical Geography. New York, 1909. Pp. 

II3-358- 

Mill, H. R., ed. The International Geography, 2d ed. New York, 1901. 

Obst, Erich. " Terminologie und Klassifikation der Berge," in Peter- 
manns Mitteilungen, lx, 1 ( 1 9 1 4) , pp . 301-310. 

Partsch, Josef. Central Europe. London, 1903. 

Peattie, Roderick. "A Geographic (Human-Use) Classification of 
Mountains," in Journal of Geography, xxxi (1932), pp. 261-264. In- 
corporated in substance in this book. 

Pirsson, L. V. Physical Geology, 2d ed. New York, 1920. Chap. xv. 
Pirsson's work is issued by the publishers as the first part, or vol- 
ume, of a two-volume set called A Text-Book of Geology. 

Reclus, Iilisee. The Earth. New York, 187 1. Pp. 11 7-16 1. 

Salisbury, R. D. Physiography, 3d ed. New York, 191 9. Pp. 31-36, 377— 
380. 

Suess, Eduard. The Face of the Earth, tr. by Hertha B. G. Sollas. Ox- 
ford, 1904-09. 4 vols. 

Supan, Alexander. Grundzuge der Physischen Erdkunde. Leipzig, 1884; 
7. Aufl., Berlin, 1927-30, 2 vols. Excellent morphogenetic classifi- 
cation. 

Upham, Warren. " A Classification of Mountain Ranges According to 
their Structure, Origin, and Age," in Appalachia, vi (1891), pp. 191- 
207. 

The Mountain Region 

Blanchard, Raoul. "The Natural Regions of the French Alps," in Geo- 
graphical Review, xi (192 1), pp. 31-49. 

Blanchard, Raoul. "Sur les noms des regions naturelles des Alpes 
franchises," in Revue de geographie alpine, xii (1924), pp. 455-462. 

Coolidge, W. A. B. The Alps in Nature and History. New York, 

i9° 8 - 
Coolidge, W. A. B. Article " Alps" in Encyclopaedia Britannica, nth and 

1 2th eds. 
Fenneman, N. M. Physiography of the Western United States. New York, 

i93i- 

Freshfield, D. W. "The Division of the Alps into Regions," in Geo- 
graphical Journal, lxxi (1928), pp. 37-42. Includes a review of the 
work of the Italian Commission. 

Krebs, Norbert. Lander kunde der Osterreichischen Alpen. 191 3. Revised 
edition, Die Ostalpen und das heutige Osterreich. Stuttgart, 1928. 2 
vols. 

Lyde, L. W. The Continent of Europe, 2d ed. London, 1924. 

Martonne, Emmanuel de. "Les divisions naturelles des Alpes," in 
Annales de giographie, xxxiv (1925), pp. 113-132. 



APPENDIX B 249 

Mar tonne, Emmanuel de. "The Carpathians: Physiographic Features 
Controlling Human Geography," in Geographical Review, iii (191 7), 
pp. 417-437. 

Newbigin, Marion Isabel. Southern Europe. London, 1932. 

Partsch, Josef. Central Europe. London, 1903. 

Piatt, J. I. Article "Alps" in Encyclopaedia Britannica, 14th ed. 

Suess, Eduard. The Face of the Earth, tr. by Hertha B. C. Sollas. Ox- 
fqrd, 1904-09. 4 vols. 



INDEX 



accessibility, 96, 1 30-1 31, 161, 181, 189, 
203-205,22*1-228. 

actionometry, 12, 14. 

adiabatic changes, 15-16, 18, 21-22. 

adret, 78, 88, 95-96, 119, 188, 190; 
bibliography, 105; see insolation, sun- 
light. 

adret, secondary, 90, 1 88. 

agriculture, importance of type, 85, 87- 
88, 159 f. 

alb, 129. 

Albania, 147, 204. 

Aletsch glacier, 48. 

aim, 129. 

alp, definition, 125-130; terms, 127- 
128; economy, 130-134. 

Alp de Loma, 181. 

alp huts, 93-94. 

alps, physical factors, 127; economy, 
130-134; bibliography, 148-151; 
ownership, 1 31-132, 167, 209, 228- 
229; see nomadism. 

Alps, the, 5, 26, 73, 86-87, l x 4> I2 °> *8i; 
sones in, 79; treeline, 11 3-1 14; sea- 
sons in, 228-232. 

Alps, French, 13, 15, 16, 24-25, 40-41, 
42-43, 64, 115, 136, 144-145. 159, 
1 60-1 61, 176, 178. 

altitude, 9, 11, 14-18, 28, 36, 85, 116, 
1 55» i^-^ 225; bibliography, 32- 
34; see zones. 

Andes, 50, 1 14. 

Andorra, 134, 204, 206, 210; zones, 81; 
bibliography, 105-106; study field 
limits, 95-96; study relief, 165-169; 
social life, 232-235. 

Appalachia, social study, 223-225. 

Appalachian Mountaineers, bibliogra- 
phy, 235-236. 

Apennines, 1 1 9. 

Ariege, 188. 

Arizona, 120. 

Arosa, 181, 190. 

atmosphere, absorbent qualities, 12, 15, 
16. 

avalanches, 48, 50, 56; track, 114; see 
snow slides. 

Aures Massif, 222. 



Austria, 93, 108, 1 13, 1 14, 153, 190, 204; 
soil temperatures, 29. 

Bagneres-de-Bigorre, 29, 67. 

balds, iio-iii. 

Balkans, transhumance, 146-148. 

Barcelonnette, 25, 90. 

barometer, 10. 

Bavaria, 44. 

beauty and health, 1 96. 

Belledonne Range, 135, 184. 

Bern, 208. 

Bolivia, 179, 180, 181, 193. 

Bora, 70-7 1 . 

boundaries, 205-209. 

Bourdillon, 7. 

Bourg-Saint-Maurice, 137. 

Bronze Age, 1 1 9. 

Carnic Alps, 94. 

Carpathian Mountains, 19, 107, 114, 
182. 

carr aires, 145. 

catastrophies, 49, 58-59, 159, 191, 216, 
229; see avalanches, floods, snow 
slides, erosion. 

Caucasus Mountains, 1 75. 

Ceillac, 138. 

Cerro de Pasco, 179-180, 191. 

Chamonix, 27. 

Champagny, 137. 

Chandolin, 141, 181. 

Chartreuse, 127, 135. 

Chaudun, 186. 

China, sacred peaks, 4. 

Chinook wind, 76. 

classification of mountains, bibliogra- 
phy, 247-248. 

climate, general references on mountain, 
bibliography, 30-3 1 . 

climatic snow line, 45-50. 

cloud pennants, 67-68. 

cloud zones, 42. 

clouds, 9, 65-66; bibliography, 77-78. 

Colorado, growing season, 23-24; zones, 

79- 
Columbia, 19. 
coma, 166. 



254 



INDEX 



communication, see transportation, ac- 
cessibility. 

Gonflent, 69-70, 135, 142, 156-157; field 
limits, 96-98; nomadism, 135. 

conservatism, 221—225. 

continentality, 48. 

Cottian Alps, 108-109. 

Courmayer, 182. 

cow rights, 132, 133. 

crop character, importance of, 85, 87- 
88, 159 f. 

cult of mountains, 5-7. 

culture gradients, 225, 228. 

Dalmatian Alps, 3 1 2-3 1 4. 

Davos, 42, 53, 64. 

deforestation, 58, 11 9-1 20, 175, 180. 

depopulation, 158-159, 175, 186-187. 

Dolomites, 94, 1 28. 

Dorfli, 186. 

Doron, 90; bibliography, see Tarentaise; 

field limits, 98-101 . 
dry holes, 43. 

dwelling, highest in Europe, 16. 
dwellings, 1 34-141, 1 81-189, 199-201. 

economic changes, 1 74. 

economic contrasts, 205. 

economic factor, field limits, 97-98, 100, 
169; forest lines, 11 7-1 22; land utili- 
zation, 158-162; population distribu- 
tion, 184-187. 

economic studies, bibliography, 1 72-1 73. 

Edelweiss, 38 

elfin wood, 1 1 o. 

Engadine, 38, 56, 119. 

erosion, 57-58, 87, 175, 186, 216. 

Ethiopia, 1 79. 

evaporation, 9, 36-38, 54, 84, 112, 114, 
128; bibliography, 59-^60. 

exposure, 9, 39, 42-43, 63, 112, 127, 128, 
188, 199; see adret, sunlight, etc. 

fatigue and mountain sickness, 192- 

193, I94-I95- 

farm limits, 85. 

field limits, 85-87; geologic factor, 93- 
94; soil factor, 94-95; economic factor, 
97-98; bibliography, 1 01 -105; in An- 
dorra, 95—96; in Gonflent, 96-98; in 
Doron, 98-101. 

forest line, 54, 108, 119, 187, 188; 
animals, 11 7-1 18; economic factors, 
1 1 7-1 22 ; evaporation, 1 1 1-1 1 2; grow- 
ing season, 11 2-1 13; mass of moun- 
tain, 1 1 3-1 1 4; snow, 1 1 4- 1 15. 



forests, importance, 120-122; bibliogra- 
phy, 122-124; height limits, 155. 
floods, 59, 121, 191. 
Foehn, 65, 72-76; mechanics of, 75. 
Freshfield, 6-7. 
Fujiyama, 3-4. 

geology, factor in field limits, 93-95. 
glaciation, 48, 86, 100-101, 130, 163, 

167; bibliography, 247. 
glaciers, 48, 54-55. 
grand montagne, 1 25. 
Graubiinden, see Grisons. 
Great Plains, 76. 
Great Smokies, 1 1 o. 
Greece, 1 19, 147. 
Grisons, 14, 59, 74, 88, 114, 158, 160, 

179, 194, 204. 
ground water, 18, 114. 
growing season, 23-25, 51, 112-113, 

"5- 

Harz Mountains, 44-45. 

haze, 195-196. 

health and altitude, 191 -197; bibliog- 
raphy, 201-202. 

health resorts, 196-197. 

heat, inherent, 16; see adiabatic changes. 

Hohen Tauren, 68, 189-190. 

house types, central Pyrenees, 142- 
144. 

human attitudes towards mountains, 
bibliography, 8. 

humidity, 9, 12,36-38. 

ice and snow slides, 57. 

industries, 51, 122, 138-139, 230—232. 

India, 179. 

Inntal, 49, 1 34. 

insolation, 9, 11-14, 49, 98, 105, 195; 
measurement, 12; bibliography, 31- 
32, 105; see also sunlight, adret, 

inversion of temperature, see tempera- 
ture. 

irrigation, see terraces. 

jasse, 142. 
Jurdes, 222. 

K 2 , 48. 

Karawanken Alps, 72, 207. 

Kashgar, 72. 

Kashmir, 204. 

Klagenfurt basin, 20. 



INDEX 



2 55 



Ladakh, 189. 

land utilization, 122; economic balance, 
152-158; in Naves, 155-158; in 
Tavetsch, 158; modern changes in, 
158-162. 

Langweis, 14. 

Leh, 12, 49. 

level land, 87, 162-165,217. 

Llosa, 80. 

Le vent, 43. 

Luz, 190. 

Marjelenalp, 142. 

mass of mountain, 18-19, 42, 48, 87, 

111,1 1 3-1 14, 128. 
Massenerhebung, 18. 
Maurienne, 144. 
mayen, 133. 
maximum precipitation and forest line, 

114. 
Metnitztal, 184. 
metric system, 233—234. 
migrations, see nomadism, transhu- 

mance. 
Mistral, 64, 70-7 1 . 

modern attitudes towards mountains, 5—7. 
Moriah, 4. 
Mont-Aimont, 138. 
Mont Blanc, 5, 42. 
Montenegro, 204. 
Mount Carmel, 84. 
Mount Canigou, 69-70. 
Mount Etna, 3-4, 193. 
Mount Everest, 69, 193. 
Mount Killington, 1 10. 
Mount Rose, 54. 
Mount Washington, 64. 
Mount Zion, 4. 
mountain, definition, 3-4. 
mountain sickness, 192-195. 
mountains, classification, bibliography, 

247-248. 
mountains, cult of, 6-7. 
mountains as barriers, 205-209. 
mountains as boundaries, 205-209. 
mountains, human attitudes towards, 

bibliography, 8. 
mountains in imagination, 4. 
mountains in middle ages, 4-5. 
mountains, origin of, bibliography, 245- 

246. 
Minister, 45. 

Nangi Parbet, 48. 

National Park of Switzerland, 28. 



Naves, land use, 1 55-1 58. 

New Mexico, 1 20. 

nomadism, 1 34-1 41, 176, 182; bibliog- 
raphy, 1 50-1 51. 

North Carolina, temperature inversions, 
20, 23. 

Oisans, 174. 

Omei, 4. 

optimum for plants, 82, 85. 

origin of mountains, bibliography, 245- 

246. 
orographic snow line, 45-50. 
Ortler Alps, 46, 48, 82, 180. 
oxygen, 193-194. 

Palestine, 84. 

Pamirs, 41. 

passes, bibliography, 236-238. 

Peru, zones, 82. 

petite montagne, 125. 

Petit St. Bernhard, 64. 

physiological limitation to population, 

191-192. 
Pic du Midi, 16, 29, 67, 68-69. 
piedmont stations, precipitation, 39. 
Pikes Peak, 3, 28. 
pla, 96, 166. 
plawuri, 147. 

political aspects, 203-2 1 7. 
political geography of mountains, bib- 
liography, 218-220. 
political philosophies of mountaineers, 

215-217. 
political unions in mountains, 2 1 0-2 1 1 . 
pollard trees, 1 20-1 2 1 . 
population distribution, and land use, 
1 59-1 61; general, 174-177; zones, 
1 77-181; details, 181-184; economic 
factors in, 184-187; bibliography, 
197-202. 
Porte de Vennasque, 1 4. 
Potosi, 191. 
Pralognan, 58, 142. 

precipitation, 38-43; and altitude, 38- 
41 ; observation inaccuracies, 38; 
zone of maximum, 41-42; bibliogra- 
phy, 60. 
pressure, 9-1 1, 14, 22, 1 93-1 94. 
Provence, 64, 71, 160. 
provincial individualism, 217* 
Py, social study, 226-227. 
Pyrenees, 25, 43, 69-70, 80, 120, 142- 
144, 160, 166, 175, 207, 216; pastoral 
buildings in, 136-138; economies in, 



256 



INDEX 



2 1 0-2 1 1 ; bibliography of political 
aspects, 220. 
Puy-dc-Dome, 60. 

Queyras, 119. 
Quito, 50. 

radiation, 29. 

rain, see precipitation. 

rain shadow, 40, 43. 

railways, 195; bibliography, 236-238. 

reflection, 87. 

Regenluft, 43. 

Regentwer, 43. 

regional studies, bibliography, 169-172. 

regions, bibliography, 248-249. 

relative humidity, 36-37, 38, 39. 

relief energy, 230-231; bibliography, 

246. 
relief in land use, 165-169. 
Rhine Valley, 191. 
Rhone delta, 149. 
Rhone Valley, 164. 
Rigi, 2 1 . 

Rincon Range, 91 . 
road building, 176, 226, 230. 
rock temperatures, 27, 28, 37. 
Rocky Mountains, 76. 
Roizonne, 187. 
run-off, 1 21-122. 

Sacred mountains, 4. 

Saint-Christophe, 50, 1 38. 

San Marino, 204. 

Santa Catalina Mountains, 24, 83. 

Santa Lucia Mountains, 222—223. 

Santis, 25. 

Savoy, 51, 134, 175; nomadism, 135. 

Schlaglawine y 56. 

seasons in the Alps, 228-232. 

S6ez, 65. 

Selkirks, 115. 

Sennhutten, 143. 

settiements, bibliography, 199-201; see 
population, dwellings, etc. 

Shelley, 6. 

Siberia, 25. 

Sicily, 87. 

Sierre Nevada (Spain), 92-93. 

Sierra Nevada (U. S. A.), 39, 40, 1 15. 

Sils, 2 1 . 

slope, 9, 11, 18, 19, 27, 39, 40, 48, 55, 
97, 114, 127,160. 

snow, 43-50; reflection of, 12; and alti- 
tude, 44; and exposure, 44; and lati- 



tude, 49; and temperature, 52; water 
content, 54; floods, 54; retention, 55; 
bibliography, 61; and forest lines, 
114-115. 

snow and human responses, 50-55. 

snow and transportation, 51-52. 

snow cover, 44-45,- 47, 52, 114, 115, 
161. 

snow line, 45-50, 1 14, 187, 188. 

snow melting, 52-53. 

snow percentages in precipitation, 9, 46. 

snow storage, 53-54. 

snow survey, 53-54. 

snow slides, 55-59; terms, 55; types, 
55-56; prevention, 56-57; prediction, 
57; examples, 57-59; effect on tem- 
perature 58; destructive force, 57-58; 
advantages, 58; and transportation, 

58. 

social character of mountain life, 221- 

235; bibliography, 238-239. 
soil, 95, 115, 116, 128; and alps, 127. 
soil erosion, see erosion, 
soil exposure, 29. 
soil in field limits, 94-95. 
soil temperature, 9, 27-29, 35, 37-38, 

87, 126, 128. 
solatia, 96. 
solane, 88. 
Sonnblick, 66. 

Spain, 1 20; see Sierra Nevada. 
St. Moritz, 181. 
Staff el, 127. 

standards of living, 225-228. 
states, mountain, 203-205. 
Stone Age, 1 1 9. 

straddle economies, 208, 2 1 0-2 1 3. 
Stubaiertal, 76, 113, 114, 127. 
sun and shade temperatures, 14. 
sunlight, 84, 88-95; bibliography, 77- 

78, 105; importance, 88-89. 
sunlight circle, 13. 
sunshine, 196, 197. 
Switzerland, 17, 45, 46, 47, 51, 57, 59, 

66, 148, 182, 190, 204, 215. 

Tarentaise, 57, 64, 68, 69, 72, 88, 127, 
136-137, 144, 160, 162; bibliography, 
106. 

Tatra, 107. 

Tavetsch, 150. 

temperature, soil, 9, 37; ranges, 9; 
sensible, 12, 13; and insolation, 14; 
and altitude, 14-18, 19; actual, 16- 
17; and latitude, 17-18; soil, 27-29; 



INDEX 



257 



bibliography, 32-34; and snow slides, 

37. 

temperature inversions, 19-23; bibliog- 
raphy, 34-35; and dwelling loca- 
tions, 188-189. 

temperature factor in zones, 84. 

term equivalents," snow slides, 55; sun 
and shade, 88; alps, 125, 129; moun- 
tain sickness, 192. 

terraces, 160, 162-165,216. 

thermal belts, see temperature inversions. 

TianShan, 41. 

Tibet, 72, 162, 1 81-182, 193, 204, 212. 

Tirol, 51, 66, 163; study of provincial- 
ism, 21 1-2 1 3. 

Tirol, South, bibliography, 220. 

topography, 11, 86, 112, 128, 189; in 
Andorra, 165-169; and settlement, 
190. 

towns, location, 189-191. 

transhumance, 144; in French Alps, 
144-146; in Balkans, 146-148. 

transhumants, 127. 

transportation, 58, 130, 159, 160, 164, 
175, 216, 222, 226, 230. 

Translyvanian Alps, 147. 

tree line, 108, 119. 

tiee requirements, 1 1 5-1 1 7. 

Trentino, see Tirol. 

Urweiden, 119. 
ubac, 88. 

Val d'Anniviers, 1 39-1 41; bibliography, 

149-150. 
umbaga, 96. 
Vald'Aoste, 182. 
Val d'Isere, 92, 153. 
Val Pellice, 1 75. 
ValRoseg, 109. 

Valais, 208, 209; bibliography, 149-150. 
Valcamonica, 183-184. 
Valespir, 69-70. 
Vallee de Conches, 45. 



Vall6e d'Ouell, 92, 95. 

valleys, shapes of alpine, bibliography, 

247. 
vapor tension, 36. 
Ventoux, 52. 
verrou t 163, 222. 
vertical temperature gradient, 16-17, 

28, 40. 
V6subie valley, 222. 
vignoble, 23. 

villages, 1 34-1 41, 1 89-1 91. 
Vintschgau, 119. 
Vlachs, 146. 
voralp, 1 33. 
Vorarlberg, 74. 
Vosges Mountains, 42. 

walls, snow slides, 56. 

waterpower, 1 21-122, 138-139. 

weathering, 28. 

Wetterhorn, 43. 

Wetterluche, 43. 

Whymper, 6. 

wind, and evaporation, 38; velocities, 
38, 63; and altitude, 38, 63-64; de- 
flection, 64, 86-87; bibliography, 77- 
78; 

wind roses, 65. 

winds, 9, 191; and mountains, 63-67; 
La Lombarde, 64-67; reverse, 66-67; 
indraft, 68-70; and tree line, 70; fall 
(Mistral and Bora), 70-71; mountain 
and valley, 71-72; Foehn, 72-77. 

Yosemite Park, 28. 
Yugoslavia, 147, 213-214. 

zero isotherm, 17. 

Zinal, 139. 

zones, 18, 19, 191; plant, 13; reversal of, 
83; value of concept, 84-85; in 
Andorra, 95-96; in the Doron, 98- 
101; bibliography, 101-106; popula- 
tion, 1 77-181.